The Approach to Internalizing Behavior, Externalizing Behavior, and Learning Difficulties of Refugee Children in Primary School: A Systematic Review
Mara Verheggen University of Amsterdam
Student number: 10272305
Assignment: Bachelor’s Thesis
Bachelor: Child Development and Education
Faculty of Social and Behavioural Sciences Assessor: dhr. dr. E. Mulder
Date: March, 2017
Table of Contents Abstract ... 3 Introduction ... 4 Method ... 9 Results ... 10 Discussion... ...19 References ... 23
Abstract
The United Nations High Commissioner for Refugees describes young asylum-seeking children as one of the most vulnerable groups of refugees. This systematic review investigated primary school-based approaches for refugee children with problems in internalizing behavior, externalizing behavior, and learning. Databases searched are Web of Science, Eric, PsychINFO, and PubMed. In total, 20 studies (altogether 1,956 children) were included. Results indicate that school-based interventions and treatments that include techniques of cognitive behavioral therapy are effective. Other forms of school-based care did not show improvements in problems of refugee children in primary school, and in a few cases problems even increased. Therefore, refugee children should be treated by evidence-based, clinical interventions conducted by professionals.
Keywords: refugee children, asylum-seeking children, primary school school-based interventions, school-based treatment,
1. Introduction
Prevailing insecurities in the world, such as the ongoing war in Syria, have caused an extreme number of 20.2 million refugees; more than half of these are children (Williams, Cassar, Siggers & Taylor, 2016). Globally, this is the highest number of refugees who have moved across borders in twenty years, doubling the number of refugees entering the European Union (Rijksoverheid, 2015). A sharp rise in the number of asylum seekers also took place in the Netherlands (COA, 2015). In November 2015, almost 50,000 people stayed in Dutch refugee centers, including more than 5,000 children aged up to twelve years. Most children of asylum seekers are from Syria (48.0%), Eritrea (10.0%), and Ethiopia (7.0%) (COA, 2015). The United Nations High Commissioner for Refugees describes young asylum-seeking children as one of the most vulnerable groups of refugees (www.unhcr.org).
Before arrival in a settlement country, children of refugee backgrounds will have experienced a wide range of stressful events. Van Os, Kalverboer, Zijlstra, Post and Knorth (2016)
investigated stressful events of refugee children in a systematic review, which included 12 studies concerning 2,585 children. Van Os and colleagues reported differences of experienced stressful events between unaccompanied children and children who are accompanied by (one of) their parents or caregivers. Many unaccompanied children reported the death of someone they really cared about (68%), drastic changes in family composition (58%), and exposure to violence (50%). The vast majority of accompanied children had lived in war conditions (89%), were exposed to war violence (80%), or were separated from close family members or lost them, had experienced torture or injury, or witnessed the killing of a close relative (35-40%).
These stressful events have a negative impact on the development of refugee children and are important risk factors for learning disabilities and behavioral and emotional problems (Hollins, King, Barnes, Papadopoulos & Grayson, 2004; Henley & Robinson, 2011; Kaplan, Stolk,
Valibhoy, Tucker & Baker, 2016, Van Os et al.). Behavioral and emotional problems in children are often divided into two major domains of dysfunctions, namely externalizing and internalizing problems (Montgomery, 2011). Both domains are related to how children deal with expressing their emotions and stress. Internalizing behaviors are negative, problematic behaviors that are mainly directed toward the self, such as anxiety, dysphoria, and depression. Children who suffer from internalizing behavior will keep emotions and stress to themselves. Externalizing behaviors are negative behaviors that are directed toward the external environment. In contrast to
internalizing behavior, feelings and thoughts are manifested outside the self in children suffering externalizing behavior. Externalizing behaviors represent conflicts with other people, with social norms, and are marked by aggression, antisocial features, defiance and impulsivity (Montgomery, 2011).
Internalizing behavior
Although there are no studies available on the current generation of refugee children, the effects of refugee experiences on children’s internalizing behavior in general are well-researched (Henley & Robinson, 2011). Earlier research has shown that there is a consensus that suggests the prevalence of internalizing behavior, primarily posttraumatic stress disorder (PTSD), depression and anxiety, is higher in refugee children than in host populations. However, the prevalence and enduring of internalizing behavior rather dependent on differences in samples, over time, and in the country of resettlement: rates of PTSD ranged from 19 to 54%, depression from 3 to 30% and anxiety from 33 to 50%. For example, Fazel, Wheeler, and Danseh (2005) determined an overall prevalence rate for PTSD of 11.0% in a systematic review identifying five surveys of 260 refugee children. In contrast, the results of Thabet and Vostanis (2003) concluded that 70.0% of Kuwati refugee children had posttraumatic stress reactions and 73.0% and 41.0% of children in the Gaza Strip suffer from mild and severe symptoms of a PTSD. Wiese and Burhorst (2007) even found a
prevalence of 80.0% in unaccompanied and accompanied children. Elevated rates of PTSD have been shown to persist up to twelve years after resettlement (Henley & Robinson, 2011). In
addition, Montgomery (2010) concluded in a nine-year follow-up study that 68.7% of the refugee children were assessed as anxious and 29% of the children were sad and depressed at arrival in the resettlement country of Denmark. After nine years, the prevalence was reduced by 33.3%, which is in accordance with other follow-up studies (Montgomery). In conclusion, it may be clear that refugee children show more internalizing behavior, mostly symptoms of PTSD, depression, and anxiety, than children in the host population.
Externalizing behavior
A few studies of refugee children focused on externalizing behavior problems.
Externalizing behavior that has been reported in studies with refugee children include aggression, hyperactivity, peer problems, and conduct disorder (Baker & Jones, 2006; Henley & Robinson, 2011). However, these studies did not report any prevalence rates. Jensen, Fjermestad, Granly and Wilhelmsen (2015) investigated 93 unaccompanied refugee children who were assessed six months after arrival in Norway. Jensen and colleagues found that 6.5% of the unaccompanied refugee children reported externalizing behavior and were indicated by the need for psychosocial intervention. Thommessen, Laghi, Cerrone, Baiocco and Todd (2013) investigated externalizing behavior among refugee children in comparison to Italian host children. Results indicated a prevalence of 20.15% among the refugee children, which was significantly higher than the host population. Nevertheless, it should be noted that Thommessen and colleagues only investigated refugee boys. Furthermore, Shaw (2003) concluded that exposure to stressful events can
Learning difficulties
Learning difficulties among refugee children have also been investigated (Henley and
Robinson, 2011). Wiebe and Burhorst found an overall prevalence of 21% of learning difficulties among refugee children. A wide range of symptoms that can occur with learning difficulties in refugees or, in most cases, traumatized children, are reported in several studies. For example, Kaplan and colleagues discuss several studies that show that stressful events in childhood are associated with learning difficulties. These difficulties include impaired memory, attention, executive skills and abstract reasoning. Furthermore, internalizing and externalizing behaviors have direct and indirect consequences on learning abilities in school. Refugee children suffer from restrictions in memory, understanding instructions, long-term memory, shifting between abstract and concrete thinking, generating problem-solving strategies, and demonstrating a solution to others (Kaplan et al. 2016). However, there is no clear, quantitative study available that highlights which symptoms of learning difficulties occur the most in refugee children. Therefore, learning difficulties will be broadly interpreted in this review.
School context
The fact that refugee children have been exposed to several stressful life events and an
accumulation of risk factors is associated with internalizing behavior, externalizing behavior, and learning difficulties, makes it necessary to examine possibilities for support and treatment upon arrival and resettlement in new countries. In addition, the need for interventions in easily
accessible environments for children is necessary, especially for newly arrived refugee children. In modern society, 90% of refugee and migrant families still do not receive the help they need (Masia-Warner, Nagle & Hansen, 2006). On one hand, these families do not have the necessary information regarding mental illness and mental health services, and on the other hand, there is a lack of knowledge among professionals in this industry. The language and culture barriers,
incomprehension, and an inadequate preparation of mental health services make it difficult to find an appropriate treatment. Therefore, it is important to consider what types of treatment are needed for refugee children (Masia-Warner, 2006).
In recent years, there has been more focus on implementation of care in school systems. For example, Fazel, Garcia and Stein (2016) interviewed 40 refugee students about impressions and experiences of mental health services integrated within school systems. In the study of Fazel and colleagues, two-thirds of the refugee children and adolescents confirmed that they preferred to be seen by professionals at school. Children experienced less stigma at school than in clinical settings and felt more comfortable when treatment took place in a safe environment (Fazel et al.). Furthermore, Masia-Warner and colleagues (2006) also suggested that mental health services should be brought into schools. According to Warner and colleagues, children spend a lot of time in school, and teachers play an important role in the contact between pupils, their parents and mental health services. Moreover, with an implementation of mental health services in schools, teachers are the primary contact for health care teams. Teachers spend more time with these children than professionals. This means that teachers who work together cooperate more with each other and hence are able to understand the performance and behavior of children in their classroom.
When help for refugee children is implemented in school systems, schools have several options. There are several school-based interventions available for refugee children, such as cognitive behavioral therapy. Schools can also adjust their surroundings to stimulate feelings of safety for refugee children. However, to date there are no reviews available that compare the possibilities of this kind of assistance for schools. Besides that, most research focuses on adolescents instead of younger refugee children. This is problematic, being that young children are the most vulnerable and also that they are still shapeable, and can quickly adapt to their new
country with the right support (Kaplan et al. 2016). Therefore, it is important to increase knowledge regarding how newly arrived children must be entitled to care and support as they enter resettlement countries.
In the context of the increasing numbers of young refugees and asylum seekers globally, the high prevalence of internalizing behavior, externalizing behavior, and learning difficulties shows that the importance of an overview of possible available care is growing. Therefore, this
systematic review aims to recognize how primary schools can support refugee children so that their internalizing behavior, externalizing behavior, and learning difficulties will be minimized.
2. Method
A multiple field study was conducted to find relevant literature in computerized databases of the Web of Science, Eric, PsychINFO, and PubMed, using the following terms in keywords, titles and abstracts: refugee children (refugee child* OR asylum-seeking child* OR war-affected child*) and school (school-based treatment OR school-based intervention* OR school
environment OR school context). Additionally, reference lists of these studies and other relevant meta-analytic reviews were checked.
Studies were selected by reading abstracts and, when necessary, the full text, relating to the following eligibility criteria. All included studies had to assess the effectiveness of a type of treatment for internalizing behavior (PTSD, anxiety, depression), externalizing behavior
(aggression, hyperactivity, peer problems, and conduct disorder) or learning difficulties (1), to include refugee children between the ages of 4 to 12 years (2), be included in a primary school-based intervention, treatment or program (3), be written in English or Dutch (4), be included a quantitative research design (5), had to be published and peer-reviewed (6) and published after 2006 (7).
Eventually, 12 studies were found that met the criteria. Because this was not enough for a proper systematic review, the criterion regarding the study design has been adapted. Five more studies with a qualitative design were included. In addition, studies including war-affected children who had lost their homes and who had the same experiences and problems as refugee children were also included. Leaving these criteria out resulted in 20 studies in total, including 1,956 children.
3. Results
Results of the literature will be sorted by type of problems that occur in refugee children, starting with internalizing behavior (1), then externalizing behavior (2), and finally learning difficulties (3). Within these sections, studies are ordered by type of intervention. Appendix A contains a table including citation, study site, country of origin, number and age in years of the participants, intervention, study design, measurements, effect sizes, and a description of the outcomes.
Internalizing behavior
Nine of the 20 studies investigated a school-based intervention on internalizing behavior. All studies researched one or a combination of the following symptoms: PTSD, anxiety and depression. Of these studies, two investigated sandplay as a school-based prevention program, five investigated a school-based intervention that included a form of CBT-techniques, and two studies investigated an alternative school-based intervention, including mediation and music therapy.
Sandplay consists of creative expression workshops and is designed for children between the ages of 4 and 6 years. At this age, children are not adequately able to express their emotions and the challenges they face. The aim of a sandplay program is to help these children gain a sense of agency through the playful transformation of this world. In the first study, Lacroix, Rousseau,
Gauthier, Singh, Giguère and Lemzoudi (2007), investigated a sandplay project in Montreal. A total of 58 refugee children from eight different countries in Asia and Africa were included. Over a period of four months, 60-minute sessions were part of the regular school day. Groups consisted of four children and one art therapist. The structure of the scenes of sand trays were analyzed using photographs and a checklist. Results showed that a sandplay program can give an
opportunity to children to express and work through their emotions and traumatic experiences. As a result, symptoms of PTSD, depression and anxiety decreased. However, no effect sizes were calculated. In a follow-up study, Rousseau, Benoit, Lacroix and Gauthier (2009) found a moderate, positive effect. In this experimental study, 105 refugee children from Pakistan, Sri Lanka, India and Bangladesh were randomized in an intervention and control group. The intervention group received 10 60-minute workshops over a period of four months, run by three art therapists. The control group followed the regular school program. The instrument used to measure internalizing behavior was the Strengths and Difficulties Questionnaire (SDQ), which was completed by parents and teachers. Effects showed a significant reduction in symptoms in the experimental group (SDQ total: t = 3.10, d=.36, p = .002; SDQ emotional symptoms (which include anxiety and depression): t = 3.35, d=.43, p = .002). In addition, teachers were very enthusiastic about the workshops and had a better understanding of the problems of their pupils. These outcomes suggested that sandplay as intervention for internalizing behavior shows
promising results.
Five studies investigated school-based cognitive behavioral therapy (CBT) for refugee children. CBT is a combination of cognitive and behavioral therapy, and the most widely used evidence-based psychological intervention. CBT focuses on cognitive restructuring and changing certain behavior. Three of five studies investigated CBT on PTSD-symptoms among children of 6 years and older and found significant effects (Beehler et al.: F(1,50) = 24.25, p<0.001;
Schottelkorb et al.: F(1,13) = 9.66, p < 0.01; Tol et al.: mean change -2.78, p < 0.001). The first study of Beehler et al. included 149 children and adolescents (49 female) from the United States, Mexico, Asia, Europe and Africa (age range from 6-21). A comprehensive Mental Health
Program for Immigrant Children, named Cultural Adjustment and Trauma Services (CATS) was implemented in two schools, one in Clifton, New Jersey, and one in Jersey City. A broad CBT-program of clinical services was included: CBT, relaxation techniques, trauma-focused cognitive-behavioral therapy (TF-CBT), supportive therapy, psycho-education, coordinating services, and family services. Clinicians and schools staff worked together as a team to identify children with difficulties. Children were involved if they identified a specific traumatic event. The client progress was measured by the CAFAS and PTSD-RI. CATS services resulted in fewer PTSD symptoms. Results of the CAFAS (p<0.01) and PTSD-RI (p=.05) were both significant. However, a limitation of the study is that it is not clear which part of the program ensured the significant results. In addition, there was no control group, and besides the use of master’s-level counseling psychologist, the study also used social workers and students who completed their practical coursework in the program. In the second study, Schottelkorb et al. (2012) investigated the effectiveness of TF-CBT in comparison with child-centered play therapy (CCPT). A total of 31 traumatized refugee children (age range from 6-13) from 15 different counties of Africa, Europe and the Middle East participated. PTSD symptoms were measured by the UCLA PTSD and the PROPS. Children were randomly assigned to the TF-CBT group or CCPT group. Children in the CCPT group received 17 sessions of 30 minutes, which is less than common practice because of limitations in the school. In addition, there were six parent consultations of 15 minutes included. Treatment in the TF-CBT group consisted of 17 sessions of 30 minutes. Parent sessions did not take place because parents were reluctant to come to the school. In the whole sample, PTSD symptoms decreased in both groups (CCPT pretest: M = 20.29 (SD = 11.87),
posttest: M = 16.93 (SD = 14.07); TF-CBT pretest: M = 22.33 (SD = 10.09), posttest: M = 20.08 (SD = 15.07)), but results for both treatments were only significant if children met criteria for full PTSD on the UCLA (p < .01, η2= .43). Limitations of the study noted by the authors were the small sample size, participants being primarily from Africa, higher rates of attrition occurring in the TF-CBT group, lack of no-intervention control group, and the use of master’s-level
counseling students. A third and final study that concerned school-based CBT was conducted by Tol et al. In central Sulawesi, 403 children of seven schools were randomized in a classroom-based intervention group (n = 237) or a wait-list group (n = 221). The intervention consisted of CBT techniques with cooperative play and creative-expressive exercises and was developed by the Center for Trauma Psychology in Boston. Children received 15 sessions in groups of 15 children over a period of five weeks by trained psychosocial counselors. Effects were calculated by the Child Posttraumatic Stress Scale and Depression Self-Rating Scale. PTSD symptoms decreased significantly in the intervention group compared to the wait-list group (PTSD symptoms change: M = -2.78 (SE = .89), p < 0.001). Complaints such as trauma idiom,
depressive symptoms and anxiety symptoms improved, but not significantly. Furthermore, the moderate decline in PTSD symptoms was only found among girls. Also in this study, effect sizes for the different parts of the intervention were again not calculated differently. However, the fourth study, conducted by Karam and colleagues (2015) found no significant effects of
classroom CBT in a study of 194 students in Lebanon (age range from 6-18). PTSD, depression and anxiety were measured by the Diagnostic Interview for Children and Adolescents (DICA-R). Children received 12 sessions of 60 minutes each in two school weeks, assessed by their teachers. No treatment effect was found (PTSD: β = .22, p = .517; depression: β = -.25, p = .501; and anxiety: β = -.44, p = .284). An explanation for these results might be that the intervention was applied by teachers who had no background in students’ internalizing problems.
Two studies combined school-based CBT with Teaching Recovery Techniques (TRT). Both studies found significant positive effects. In the first study of Barron, Abdullah and Smith (2012), school counselors were trained in the program. A total of 40 Palestinian children (age range from 11-14) participated in the intervention group (n = 28) or wait-list group (n = 22). All had PTSD symptoms, which were measured by the UCLA and CRIES-13 questionnaires. The intervention group received five sessions of 90 minutes in a time period of five weeks. Large effect sizes were found for program impact on reducing PTSD (d = .76) and depression (d = 1.24). Furthermore, in the second study that combines school-based CBT with TRT, Ooi and colleagues (2016) did a pre-test, post-test, and three-month follow-up. A total of 82 participants were recruited and outcome measures were, as in the study of Barron and colleagues, measured by UCLA, CRIES-13 and additionally the SDQ. The same results were found. Children in the intervention condition experienced, overall, a greater symptom reduction in problems than participants in the wait-list condition at post-test (F (1, 155) = 5.20, p = .024, η2 = .07), and after the three-month follow-up (F (2, 122) = 7.24, p = .001, η2 = .20). Internalizing problems,
including PTSD and depression, decreased significantly (t(121) = 2.47, p=.015). In addition, Ooi and colleagues also investigated externalizing problems, which will be discussed later. The authors of both articles concluded that the intervention was effective and that school-based CBT will become more similar in future. Because four of five studies found significant results, it can be concluded that a school-based treatment for internalizing problems in refugee children is effective if it includes a form of CBT. However, it should be noted that all studies used a broad program of which CBT was only one component, and that the effect sizes could be a result of the cumulative effect of the different components of which the treatment consisted.
Furthermore, two studies investigated alternative school-based interventions in internalizing behavior. First, Peltronen, Qouta, El Sarraj and Punamäki (2012) researched the
effectiveness of school-based mediation intervention on internalizing behavior and externalizing behavior. The study included 225 Palestinian children (age range from 10-14). Internalizing behavior, PTSD symptoms (F (1, 149) = 7.70, p < .006, η2= .003) and depressive symptoms (F (1, 149) = 5.21, p < .024, η2= .03) increased significantly. Results of externalizing behavior will be discussed in the next section. In the second study of Backer and Jones (2006), the effect of music therapy services on internalizing behavior, externalizing behavior, and learning difficulties was researched. Refugee children of Sudan, Iran, Liberia, Rwanda and Ethiopia (N = 31) were included in a pilot study in an intensive English-language center for newly arrived immigrant and refugee youth in Queensland. Students received 20 sessions of 30-40 minutes in 20 weeks. The Behavior Assessment System for Children (BASC) was used to measure internalizing behavior, externalizing behavior and learning difficulties, and was evaluated by teachers. After 20 weeks, internalizing problems (F = 3.73, p = .02) increased significantly. Results of externalizing behavior and learning difficulties are presented in the following sections. Furthermore, it may be clear that these alternative therapies are not effective in decreasing internalizing problems for refugee children.
Externalizing behavior
Five of the 20 studies investigated a school-based intervention on externalizing behavior. All studies researched one or a combination of the following symptoms: aggression,
hyperactivity, peer problems, and conduct disorder. Of these studies, one investigated the effects of CBT on externalizing problems, two investigated alternative therapies, and two studies implemented a mental health service in school.
The first three studies that investigated the effects of school-based interventions or programs are discussed above. First, in the study of Ooi and colleagues, which included a school-based CBT with TRT, symptoms of externalizing behavior decreased significantly (t(122) = 12.51,
p<.001). In addition, in the study of Peltronen and colleagues, internalizing behavior increased significantly, but there was no change found in externalizing behavior, including aggression and peer problems (measured by friendship quality and prosocial behavior). Third, in the study of Baker and Jones, who researched the effect of music therapy, externalizing problems increased significantly (F = 4,92; p =.01).
Two other studies implemented a mental health service in schools to decrease
externalizing behavior. In the first study, Fazel, Doll and Stein (2009) conducted further research to recognize externalized behavior in refugee children. In their study, which included 47 children (age range from 4-19), the authors implemented a school-based mental health service including family work, individual therapy, and group work. The core activity of the service consisted of a consultation with the mental health key worker and teacher every week. Two primary schools and one secondary school with refugee children from the Balkans, Asia, India and Africa participated. To create a control group, children in the treatment group were compared with non-refugee children of an ethnic minority group. Measurements of externalizing behavior were calculated with the Strengths and Difficulties Questionnaire and completed by teachers. While the refugee children continued to have a significantly higher SDQ total score than those in the control group at the end of the school year, the total SDQ score in all groups decreased significantly (F (1, 138) = 5.9, p = .016). The highest changes were observed in the peer problems scale (F (1, 138) = 8.1, p = .005) and hyperactivity scale (F (1, 138) = 3.9, p = .002). However, also in this study, the sample size was small and 38% of students were attending secondary school. In the second study by Durà-Vilà (2012), a similar approach was taken. Services on site following consultation with staff, clinical interviews, and other assessments with the family were provided. Children
demonstrated significant improvement in their overall SDQ scores (t = 2.787, p = .010), hyperactivity scores (t = 2.635, p = .015) and peer problem scores (t = 2.551, p = .017).
Therefore, it can be concluded that these mental health services are effective in decreasing externalizing behavior.
Learning difficulties
Seven of the 20 studies investigated a school-based intervention on learning difficulties. Learning difficulties were researched as school problems on one hand, whereas other studies investigated how to support the school climate to minimalize learning difficulties.
The first two studies are discussed in the previous section. First, in the study of Barron and colleagues, a small effect size was found for reducing the impact of trauma on learning difficulties by CBT techniques (d = .35). Second, Baker and Jones, who researched the effect of music therapy, found in their research that school problems increased significantly (F = 6.28, p = .00). In this study, externalizing behavior also increased, and teachers reported that the
externalizing problems led to more learning difficulties for the refugee students and for the students and teachers around them.
In the third study, Ager and colleagues (2011) investigated the impact of the school-based psychosocial structured activities (PSSA) program on children who have undergone psychosocial stress during war in Uganda. Children received 15 thematic and closing activities sessions of 60 minutes given by trained school teachers. Themes during sessions were, for example, reactions regarding danger and coping skills, to improve the interests in school, responsibility and hardworking. Brief ethnographic interviews were conducted to determine the well-being of children in school. Results were calculated by a coding system. After a follow-up period of 12 months, the well-being scores significantly increased over time for children receiving the intervention (treatment group: b = 5.4, p < .001, control group: b = 4.0, p = .01).
In the fourth study, Dunn, Bundy and Woodrow (2012) researched the effect of drama pedagogy with digital technologies to support learning in refugee children. The aim of the
program was to empower the children in solving problems and support the learning context. Because of the study design, no conclusions can be made. However, analysis showed that technologies used within drama therapy served several key functions related to language development, information provision, narrative development, identification and opportunities to share experiences with classmates and teachers.
Three qualitative studies investigated how schools, classroom teachers and staff can support refugee children. Due, Riggs and Augoustinos (2016) did qualitative research in combination with an ethnographic data analysis in three Australian primary schools, including migrant and refugee children from 20 different countries with 12 different cultural and linguistic backgrounds. Due and colleagues concluded that refugee children in classrooms should include opportunities and challenges for teachers and other children in the class. The main opportunities were that diverse classrooms offer the possibility to create spaces for children to share their experiences and knowledge (1); and classrooms can foster positive interactions between students (2). Because all children had different backgrounds, the inclusivity was maximized. Two
challenges consisted of the high rates of distress that refugee children experienced in classrooms and tension between teachers and children, because children had no English knowledge, no prior schooling and were suffering from trauma. In addition, Due and Riggs (2016) implemented a photo elicitation and interviews in three primary schools in South Australia. A total of 63 migrant and refugee children were interviewed about the pictures they made during the intervention. In conclusion, children felt safe in school when there was a caring environment in which their identities and experiences were reflected. In addition, refugee children felt safe when they had a good relationship with their classroom teacher and other staff. Limitations of this study were the use of small sample sizes, not taking into account the age of students, and the definitions of care and support overlapped during interviews. In a third study of Block, Cross, Riggs and Gibbs
(2014), the School Support Program was evaluated in 21 primary schools, with the aim to establish a more inclusive environment in schools to support refugee children and their families. Evaluation took place by conducting semi-structured interviews among teachers and school staff. The authors concluded that teachers feel more confident and less anxious when they have
knowledge of the background of refugee children. However, active and strong leadership of teachers provided more effective support for children of a refugee background. In addition to effective support, teachers need an understanding of the complex background and the needs of refugee students.
In the area of learning difficulties, the intervention that included CBT techniques is again effective. In addition, the PSSA program also presented significant results. Furthermore,
qualitative studies concluded that a safe environment and good relationships with peers and classroom teachers is important for refugee children.
4. Discussion
Young asylum-seeking children are one of the most vulnerable groups of refugees. Therefore, this systematic review had the aim to give an overview of literature to increase knowledge in which way newly arrived children could be entitled to care and support as they enter resettlement countries. The fact that refugee children have been exposed to several stressful life events and an accumulation of risk factors is associated with internalizing behavior,
externalizing behavior and learning difficulties. Primary schools can play an important role in implementing care for refugee children, because schools ensure a safe environment for children and can contact professionals and parents. In conclusion, the only thing that has been identified with certainly, is that effective school-based interventions for internalizing behavior,
externalizing behavior and learning problems include all a form of CBT. In case of externalizing behavior, the PSSA program also showed significant results. Furthermore, other school-based
interventions do not show a decrease in problems, and in two studies internalizing behavior, externalizing behavior and learning difficulties of refugee children even increased.
With the results of this systematic review, the question arises regarding why CBT is more successful than other (parts of) interventions. First, CBT is the most investigated treatment in the psychology. In addition, CBT includes a broad approach that focuses on both cognition and behavior and is therefore applicable in many conditions (Bögels & Van Oppen, 2011). With the cognitive aspect in CBT, therapists try to change dysfunctional schemes that are responsible for many psychological problems and diseases (Van Deth, 2014). A scheme is a behavior pattern that people use to interpret different situations and circumstances. Therapists try to change the
schemes with the behavioral aspect of CBT. The mean idea is that if cognitive schemes are no longer dysfunctional, a person is able to deal with different situations (Van Deth, 2014). Therefore, CBT focuses on the underlying causes instead of the outcomes that other therapies, such as school mediation therapy and music therapy, do.
The strength of this study is that it is the first one that provides an overview of the currently available interventions and approaches to primary schools with refugee children. It shows that the available options for schools are minimal, and that there is a lack of knowledge of dealing with the complex backgrounds of refuge children in the age of 4 to 12 years. Except of school-based CBT, there is no effective intervention. However, school can play a key role in treatment of refugee children (Fazel, Garcia & Stein, 2016). Teachers play an important role in supporting and mediating contact with mental health services, because this in particular can be difficult for newly arrived refugees. In addition, teachers’ screening may identify children not receiving service, and are also the ones who could contact parents and professional caregivers.
This review also has several limitations. First, the design quality of the included studies was restricted. Except of three studies, all have small samples sizes consisting of less than
hundred children. Studies with small sample sizes tend to have larger effect sizes than those with larger samples (Slavin & Smith, 2009). In addition, there was no randomized (wait-list) control group in nineteen studies, which makes it difficult to compare the effect of interventions. A control group is one of the most important components in testing the effectiveness of
interventions, it makes it possible to eliminate the influence of confounding variables, bias and coincidences (Agresti & Franklin, 2013). Furthermore, generalizability of results is limited to the current population of refugee children because all studies included different groups of refugees of previous flows. For example, 97% of Syrian children of primary school age were attending school before March 2011 (UNESCO, 2013). In contrast to Eritrea, where the education level is generally low (UNESCO, 2012). Kaplan and colleagues (2016) noticed that the amount and quality of education the child has received are important predictors of the cognitive development, academically performance and language acquisition when resettling in new countries. Therefore, larger sample sizes and a diverse refugee samples are recommended in future research.
Second, in many studies the treatment integrity was low and the interventions were carried out by students instead of licensed mental health professionals. Research has shown that effective interventions for behavior problems and mental health have to apply to overall active factors, regardless of the problems that are being addressed. Garland, Hawley, Brookman-Frazee and Hurlburg (2008) concluded that interventions that are being performed by professionals are more effective than interventions in which this is not the case. In addition, interventions have better results when the adequacy and quality of the execution is monitored. In the future, only professionals should be performing in research on refugee children.
Finally, this systematic review ends on the note that researchers should be careful with this vulnerable group of children. As mentioned by Peltronen and colleagues (2012), carefulness is required when planning interventions for refugee children. Many discussed interventions are
not powerful enough to help children in a complex situation which is unfamiliar and difficult to assess. According to articles 24 and 39 of the Convention on the Rights of the Child, children are entitled to the best possible health and to the best available health care, which has to be accessible to all children. In addition, states are obliged to create an environment in which recovery and reintegration of refugee children is centered and is conducive to the health, self-respect and dignity of these children. It is in breach of these rights and of concern that interventions are carried out in which the symptoms of refugee children increased significantly. Therefore, refugee children should be treated by evidence-based, clinical interventions conducted by professionals.
References
*Ager, A., Akesson, B., Stark, L., Flouri, E., Okot, B., McCollister, F., & Boothby, N. (2011). The impact of the school‐based Psychosocial Structured Activities (PSSA) program on conflict‐affected children in northern Uganda. Journal of Child Psychology and
Psychiatry, 52(11), 1124-1133. doi:10.1111/j.1469-7610.2011.02407.x
Agresti, A., & Franklin, C. (2007). The art and science of learning from data. Upper Saddle River, NJ: Prentice Hall.
*Baker, F., & Jones, C. (2006). The effect of music therapy services on classroom behaviours of newly arrived refugee students in Australia—a pilot study. Emotional and Behavioural Difficulties, 11(4), 249-260. doi:10.1080/13632750601022170
*Barron, I. G., Abdallah, G., & Smith, P. (2013). Randomized control trial of a CBT trauma recovery program in Palestinian schools. Journal of Loss and Trauma, 18(4), 306-321. doi:10.1080/15325024.2012.688712
*Beehler, S., Birman, D., & Campbell, R. (2012). The effectiveness of cultural adjustment and trauma services (CATS): Generating practice-based evidence on a comprehensive, school-based mental health intervention for immigrant youth. American journal of community psychology, 50(1-2), 155-168. doi:10.1007/s10464-011-9486-2
*Berger, R., Pat‐Horenczyk, R., & Gelkopf, M. (2007). School‐based intervention for prevention and treatment of elementary‐students' terror‐related distress in Israel: A quasi‐randomized controlled trial. Journal of traumatic stress, 20(4), 541-551. doi:10.1002/jts.20225
*Block, K., Cross, S., Riggs, E., & Gibbs, L. (2014). Supporting schools to create an inclusive environment for refugee students. International Journal of Inclusive Education, 18(12), 1337-1355. doi:10.1080/13603116.2014.899636
*Catani, C., Kohiladevy, M., Ruf, M., Schauer, E., Elbert, T., & Neuner, F. (2009). Treating children traumatized by war and Tsunami: a comparison between exposure therapy and meditation-relaxation in North-East Sri Lanka. BMC psychiatry, 9(1), 22.
doi:10.1186/1471-244X-9-22
COA. (2015). Cijfers. Retrieved from: https://www.coa.nl/nl/over-coa/cijfers
*Due, C., & Riggs, D. W. (2016). Care for children with migrant or refugee backgrounds in the school context. Children Australia, 41(03), 190-200. doi:10.1017/cha.2016.24
*Due, C., Riggs, D. W., & Augoustinos, M. (2016). Experiences of school belonging for young children with refugee backgrounds. The Educational and Developmental
Psychologist, 33(01), 33-53. doi:10.1017/edp.2016.9
*Dunn, J., Bundy, P., & Woodrow, N. (2012). Combining drama pedagogy with digital technologies to support the language learning needs of newly arrived refugee children: a classroom case study. Research in Drama Education: The Journal of Applied Theatre and Performance, 17(4), 477-499. doi:10.1080/13569783.2012.727622
*Durà-Vilà, G., Klasen, H., Makatini, Z., & Hodes, M. (2012). Mental health problems of young refugees: Duration of settlement, risk factors and community-based interventions. Clinical Child Psychology and Psychiatry, 18(4). doi:10.1177/1359104512462549
Ep-Nuffic (2015). Onderwijssysteem Syrië. Het Syrische onderwijssysteem beschreven en vergeleken met het Nederlandse.
https://www.epnuffic.nl/documentatie/vind-eenpublicatie/onderwijssysteem-syrie.pdf
*Fazel, M., Doll, H., & Stein, A. (2009). A school-based mental health intervention for refugee children: An exploratory study. Clinical Child Psychology and Psychiatry, 14(2), 297-309. doi:10.1177/1359104508100128
Fazel, M., Garcia, J., & Stein, A. (2016). The right location? Experiences of refugee adolescents seen by school-based mental health services. Clinical child psychology and
psychiatry, 21(3), 368-380.Fazel, M., Wheeler, J., & Danesh, J. (2005). Prevalence of serious mental disorder in 7000 refugees resettled in western countries: a systematic review. The Lancet, 365(9467), 1309-1314. doi:10.1016/S0140-6736(05)61027-6 Garland, A.F., Hawley, K.M., Brookman-Frazee, L., & Hurlburt, M.S. (2008). Identifying
Common Elements of Evidence-Based Psychosocial Treatments for Children’s Disruptive Behavior Problems. Journal of the American Academy of child and Adolescent Psychiatry, 47, 505-514. doi:10.1097/CHI.0b013e31816765c2
Henley, J., & Robinson, J. (2011). Mental health issues among refugee children and
adolescents. Clinical Psychologist, 15(2), 51-62. doi:10.1111/j.1742-9552.2011.00024.x Jensen, T. K., Fjermestad, K. W., Granly, L., & Wilhelmsen, N. H. (2015). Stressful life
experiences and mental health problems among unaccompanied asylum-seeking children. Clinical child psychology and psychiatry, 20(1),
106-116. doi:10.1177/1359104513499356
Kaplan, I., Stolk, Y., Valibhoy, M., Tucker, A., & Baker, J. (2016). Cognitive assessment of refugee children: Effects of trauma and new language acquisition. Transcultural psychiatry, 53(1), 81-109. doi:10.1177/1363461515612933
*Karam, E. G., Fayyad, J., Karam, A. N., Tabet, C. C., Melhem, N., Mneimneh, Z., & Dimassi, H. (2008). Effectiveness and specificity of a classroom‐based group intervention in children and adolescents exposed to war in Lebanon. World Psychiatry, 7(2), 103-109.
*Lacroix, L., Rousseau, C., Gauthier, M. F., Singh, A., Giguère, N., & Lemzoudi, Y. (2007). Immigrant and refugee preschoolers’ sandplay representations of the tsunami. The Arts in Psychotherapy, 34(2), 99-113. doi:10.1016/j.aip.2006.09.006
Masia-Warner, C., Nagle, D., & Hansen, D. (2006). Bringing evidence-based child mental health services to the schools: General issues and specific populations. Education & Treatment of Children, 29(2), 165–172
Montgomery, E. (2010). Trauma and resilience in young refugees: A 9-year follow-up study. Development and psychopathology, 22(02), 477-489.
doi:10.1017/S0954579410000180
*Ooi, C. S., Rooney, R. M., Roberts, C., Kane, R. T., Wright, B., & Chatzisarantis, N. (2016). The Efficacy of a Group Cognitive Behavioral Therapy for War-Affected Young Migrants Living in Australia: A Cluster Randomized Controlled Trial. Frontiers in Psychology, 7. doi:10.3389/fpsyg.2016.01641
*Peltonen K, Qouta S, El Sarraj E, Punamaki RL. (2012). Effectiveness of school-based intervention in enhancing mental health and social functioning among war-affected children. Traumatology, 37, 37-46. doi:10.1177/1534765612437380
*Riggs, D. W., & Due, C. (2011). (Un) common ground?: English language acquisition and experiences of exclusion amongst new arrival students in South Australian primary schools. Identities, 18(3), 273-290. doi:10.1080/1070289X.2011.635373
*Rousseau, C., Benoit, M., Lacroix, L., & Gauthier, M. F. (2009). Evaluation of a sandplay program for preschoolers in a multiethnic neighborhood. Journal of child psychology and psychiatry, 50(6), 743-750. doi:10.1111/j.1469-7610.2008.02003.x
*Schottelkorb, A. A., Doumas, D. M., & Garcia, R. (2012). Treatment for childhood refugee trauma: A randomized, controlled trial. International Journal of Play Therapy, 21(2), 57-73. doi:10.1037/a0027430
Slavin, R., & Smith, D. (2009). The relationship between sample sizes and effect sizes in systematic reviews in education. Educational Evaluation and Policy Analysis, 31(4), 500-506. doi:10.3102/0162373709352369
Thabet, A. A. M., Abed, Y., & Vostanis, P. (2004). Comorbidity of PTSD and depression among refugee children during war conflict. Journal of Child Psychology and Psychiatry, 45(3), 533-542. doi:10.1111/j.1469-7610.2004.00243.x
Thommessen, S., Laghi, F., Cerrone, C., Baiocco, R., & Todd, B. K. (2013). Internalizing and externalizing symptoms among unaccompanied refugee and Italian adolescents. Children
and Youth Services Review, 35(1), 7-10. doi:10.1016/j.childyouth.2012.10.007
*Tol, W. A., Komproe, I. H., Susanty, D., Jordans, M. J., Macy, R. D., & De Jong, J. T. (2008). School-based mental health intervention for children affected by political violence in Indonesia: a cluster randomized trial. Jama, 300(6), 655-662. doi:10.1001/jama.300.6.655 UNESCO, 2012 High level International Round Table on Literacy: “Reaching the 2015 Literacy
Target: Delivering on the promise”.
http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/ED/pdf/Eritrea.pdf
UNICEF, World Vision, UNHCR, Save the Children (2013), Syria Crisis: Education Interrupted. Global action to rescue the schooling of a generation.
http://www.unicef.org/media/files/Education_Interrupted_Dec_2013.pdf
Van Deth, R. (2014). Psychotherapie: Van theorie tot praktijk (3rd rev. ed.). Houten, The Netherlands: Bohn Stafleu van Loghum.
Van Os, E. C. C., Kalverboer, M. E., Zijlstra, A. E., Post, W. J., & Knorth, E. J. (2016). Knowledge of the Unknown Child: A Systematic Review of the Elements of the Best Interests of the Child Assessment for Recently Arrived Refugee Children. Clinical Child and Family Psychology Review, 19(3), 185-203. doi:10.1007/s10567-016-0209-y
Wiese, E. B. P., & Burhorst, I. (2007). The mental health of asylum-seeking and refugee children and adolescents attending a clinic in the Netherlands. Transcultural Psychiatry, 44(4), 596-613.
Williams, B., Cassar, C., Siggers, G., & Taylor, S. (2016). Medical and social issues of child refugees in Europe. Archives of Disease in Childhood. doi:10.1136/archdischild-2016-310657
Appendix: Characteristics of included studies Table 1
Characteristics of Included Studies Citation Study site Country
of origin Number of participants Age in years Child outcomes Intervention Study Design
Measurements Effect Sizes Conclusion
Ager et al.
(2011) Uganda Uganda 191 7-12 Learning Creative arts (PSSA) RCT Modified BEI β = 5.4, p < .001 Effective Baker et al. (2006) Australia Sudan, Iran, Liberia, Rwanda, Ethiopia 31 11-16 Externalizing Internalizing Learning Music Therapy Cross-over design BASC Externalizing problems (F = 4,92; p =.01), Internalizing problems (F = 3.73, p = .02), Learning difficulties (F = 6.28, p = .00) Not effective
Barron et al. (2012)
Palestine Palestine 40 11-14 Internalizing CBT- Trauma- Recovery-Program Experimen tal SDQ PTSD (d=.76); depression (d=1.24); dental health difficulties (d=.90); traumatic grief (d=.96); reducing impact trauma (d=.35) Effective Beehler et al. (2011) USA US, Mexico, Asia, Europe, Africa 149 6-21 Internalizing (PTSD) Comprehens ive school-based mental health service model Quasi-experimen tal CAFAS, PTSD-RI Cumulative totals of supportive therapy F(1, 150) =24.25, p<0.001 TF-CBT = F(1, 150) =7.86, p<0.01 Effective Berger et al. (2007)
Israel Israel 142 7-11 Internalizing CBT Quasi-randomize d controlled trial UCLA PTSD, PTSD d = .99, somatic complains d = .64, separation anxiety d =.55, general anxiety d = .96 Effective SCARED, Child Diagnostic Interview Schedule
Block et al. (2014) Australia Middle East, Burma, Sri Lanka, Africa 21 primary schools
4-12 Learning The School Support Program Qualitative VicHealth checklist for partnership - Hard to tell Catani et
al. (2009) Germany Sri-Lanka 31 8-14 Internalizing Narrative Exposure Therapy for Children; meditation-relaxation Qualitative UCLA PTSD Index for DSM-IV - Hard to tell Due et al. (2016) Australia 20 different countries 63 5-13 Learning Photo elicitation methodolog y, interviews
Qualitative - - Hard to tell
Due et al. (2016) Australia 20 different countries 4-12 Learning Ethnographi c data analysis, interviews
Qualitative - - Hard to tell
Dunn et al. (2012)
Australia Burundi 15 8-12 Learning Drama Therapy
Dura-Vila et al. (2013) UK Balkan, Asia, China 74 3-17 Externalizing Support therapy Cohort study SDQ SDQ scores (t = 2.787, p = .010), hyperactivity scores (t = 2.635, p = .015) and peer problem scores (t = 2.551, p = .017). Effective Fazel et
al. (2009) UK Balkans, Asia, India, Africa
47 4-19 Externalizing The school-based mental health service Quasi-Experimen tal
SDQ (teacher) Overall p=0.003; Effective
Hyperactivity p=0.069 Conduct disorder p=.743 Peer problems p=0.005;
Karam et al. (2008)
Lebanon Lebanon 194 6-18 Internalizing Classroom CBT Quasi-experimen tal DICA-R PTSD: β = .22, p = .517; depression: β = -.25, p = .501; and anxiety: β = -.44, p = .284 Not effective Lacroix et al. (2007) Canada Pakistan, Sri Lanka, India, Banglades h, Ghana, Morocco, Tunisia, Haiti 58 4-6 Internalizing Sandplay program
Qualitative - - Hard to tell
Ooi et al.
(2016) Australia - 82 10-17 Internalizing Externalizing Teaching Recovery Techniques RCT CRIES-13 Internalizing t(121)=2.47, p=.015, externalizing t(122)=12.51, p<.001 Effective DSRS SDQ UCLA-PTSD Peltonen et al. (2012)
Finland Palestine 225 10-14 Internalizing School mediation intervention
Quasi-experimen tal
25-item list for military trauma; CPTS-RI; CDI; SDQ; Friendship Quality Scale, MAQ PTSD symptoms p<.006 Not effective Depression p<.024
Riggs et al. (2011)
Australia - 75 4-12 New Arrival
Programme Qualitative Ethnography, observation, field notes, NAP-questionnaire - Hard to tell Rousseau et al. (2009) Canada Pakistan, Sri Lanka, India, Banglades h 105 4-6 Internalizing Externalizing Sandplay program Quasi-experimen tal SDQ (parents and teachers) SDQ total score parents’ report: t=2.49, p=.014; SDQ total score teachers’ report: t=1.99, p=.05 Promising results Schottelk orb et al. (2012) US Afghanist an, Bosnia, Burma, Burundi, Congo, Iraq, Kenya, Liberia, Nepal, Russia, Rwanda, Somalia, Tanzania, Turkey, Uzbekista n 31 6-13 Internalizing Child-centered play therapy, trauma-focused CBT Quasi-experimen tal UCLA PTSD Index for DSM-IV; PROPS PTSD Time: Wilks’ λ = .57, F(1, 13) = 9.66, p > .01, η2 = .43 Time x Group: Wilks’ λ = .99, F(1, 13) = 0.18, p = .68, η2 = .01 Effective
Tol et al. (2008) Indonesia Indonesia 403 M=9. 94 (SD= 1.21) Internalizing Classroom-based intervention integrating CBT techniques, cooperative play and creative-expressive excercises Quasi-experimen tal Child Post-Traumatic Stress Scale; Depression Self-Rating Scale; Self-Report for Anxiety Related Disorders, PTSD symptoms 2.78(.89), p<.001; Depressive Symptoms .70 (.40), p=.02; Anxiety Symptoms .12 (.22), p=.15; Effective
