Mental health problems in refugee and immigrant primary school children in Flanders,Belgium

Abstract

Background

European countries face the challenge of promoting refugee and immigrant children’s well-being within their host communities, invoking the necessity of adequate mental health assessment. This study aims to contribute to document the psychosocial well-being of primary school refugee and non-refugee immigrant children in Flanders, Belgium.

Method

A total of 120 children (8–12 years old) with migration backgrounds participated in the study. Through self-report, parent and teacher questionnaires we scrutinized externalizing and internalizing behavioral problems, post-traumatic stress problems, and classroom relationships.

Results

Thirty percent of the participants reported high levels of post-traumatic stress; around 25% reported a high or very high prevalence of internalizing and externalizing behavioral problems. Self-reported mental health problems are elevated in comparison to the general population. Refugee children did not report more difficulties than their immigrant peers. In the perception of parents and teachers, respectively 20% and 5% of children showed high or very high amounts of internalizing and externalizing behavioral difficulties. Almost 70% of the participants perceived the class climate as unsafe.

Conclusions

Refugee and immigrant children are at risk for mental health difficulties, and experience classroom dynamics as markedly distressful. School-based intervention might be particularly suited to support these children’s psychosocial well-being in resettlement.

Keywords

refugee children immigrant children mental health school well-being

Introduction

European societies face the challenge of promoting refugee and non-refugee immigrant children’s adaptation within their host communities. Indeed, studies document how migration histories and resettlement conditions render these children at an increased risk for mental health difficulties. For refugee children, pre-migratory stressors of collective violence as well as flight-related stressors of family separation and deprivation are compounded by post-flight stressors of cultural change, residence insecurity, isolation and experiences of discrimination (Spaas et al., 2022). These cumulative stressors may also impact affective security within refugee family relationships and invoke family conflict and decreased parental availability (De Haene et al., 2010b). Studies indicate how refugee children’s mental health may be at-risk in the face of this cumulation of disruptive life-experiences and document high prevalence of post-traumatic stress disorder (PTSD), depression, and internalizing and externalizing problems (e.g., Hodes & Vostanis, 2019).

For non-refugee immigrant children (referred to as immigrant children in what follows), studies consistently show that migration status is associated to at-risk mental health functioning, particular internalizing problems (for review, see Kouider et al., 2014). Although resettlement stressors of acculturation, discrimination, or deprivation are contextual factors immigrant children share with refugee children, it remains unclear whether immigrant children’s increased mental health vulnerability shows analogous profiles as compared to refugee children. Here, the paucity of evidence is equivocal: while a US-based study by Betancourt et al. (2017) documented higher levels of trauma exposure and post-trauma symptoms in refugee than in immigrant children, another US study showed high levels of trauma exposure (including domestic and community violence) in recent (non-refugee) immigrant children (Jaycox et al., 2002).

With studies indicating psychosocial vulnerability in refugee and immigrant children, it seems relevant to explore mental health profiles in both groups in which migration histories may comprise of shared and non-shared stressors. In exploring mental health profiles in both groups, an exploration of school-based social relationships as complementary outcome variable seems relevant. Studies demonstrate the role of school-related variables (e.g., positive peer and teacher relations, school belonging) in determining young refugees’ and immigrants’ mental health. For refugee children, positive peer and teacher-pupil relationships are associated with improved psychological functioning, being important sources of support in coping with distress and social exclusion or discrimination (Fazel, 2015). School-related factors like school belonging and sense of safety at school have been found to protect against psychiatric disorders such as PTSD, depression and anxiety (e.g., Due et al., 2016). Similarly, for immigrant children, studies highlight the detrimental impact of exclusion and discrimination, and document the protective role of positive peer relationships and sense of belonging in the school context (Gharaei et al., 2019). These findings in both refugee and immigrant children indicate how school-based social relationships may operate as both a buffer against the psychological sequelae of migration-related stressors or as risk factor for mental health problems. Therefore, assessing mental health profiles in the context of school and including an exploration of their sense of belonging and safety in the classroom peer context seems relevant.

In this study, we aim to investigate mental health profiles in refugee and immigrant children in the primary school context, including an exploratory assessment of class well-being and social relationships. As existing studies for the Belgian context have focused on adolescent refugees and immigrants (e.g., Derluyn et al., 2008; Vervliet et al., 2014), this study provides a first systematic assessment of mental health in primary school-aged refugee and immigrant children.

Methodology

Study setting and participants

This study is part of a larger research project on the impact of a creative arts-based expression program (Rousseau et al., 2004) on refugee children’s mental health and peer relationships. The study reports on the baseline assessment of mental health problems and classroom relationships.

Purposive recruitment of schools was conducted in 2017–2018 in the Flemish community of Belgium. Targeting schools with a high number of refugee and immigrant pupils, 30 schools were contacted through the Agency of Educational Services (AGODI); three schools were selected, based on their multi-ethnic composition and presence of refugee pupils, interest in study participation, expected pupil drop-out rates (e.g., asylum decisions), and geographical location. The schools were located in one Flemish multi-ethnic city, which increased homogeneity in participants’ social environment and ensured accessibility for data collection and intervention implementation. Pupils in the third to sixth grade (aged 8–12) were eligible for participation, in line with the target group of the intervention under study and the skills needed to complete self-report questionnaires.

Children were included in the study when their parents provided active informed consent: following a collective information session including visual and interpreter support, parents received oral and written information about the study and could sign consent. Through close collaboration with school teams, we engaged in an intensive process of outreach to parents, for example, addressing them at parent-teacher meetings or at the school gate. Out of the 274 children approached for participation, parental informed consent was obtained for 120 of them. This 44% response rate mirrors the complexities of access and alliance-building in research with refugee communities (De Haene et al., 2010a; de Smet et al., 2020).

Following informed consent, parents completed standardized questionnaires on background characteristics (e.g., child gender, child age, children’s and parents’ country of origin, time in Belgium, parental education, legal status). Based on these background characteristics, participants were categorized as refugee children or immigrant children. Life-trajectories of forced displacement induced by persecution, war, or violence were a precondition to be categorized as refugees. The country of origin (e.g., Syria, Afghanistan) and legal status (e.g., refugee, asylum seeker) informed the categorization as refugees, in unclear cases complemented by an oral exploration with parents. Fifty-five participants were categorized as refugee children, 62 as immigrant children, and for three participants information that would allow categorization was lacking.

Table 1 summarizes sample demographic characteristics. The participant sample was heterogeneous in terms of countries of origin, with participating children being born in 32 different countries. Refugee children were mainly born in Syria (N = 15, 27.3%), Somalia (N = 11, 20%), Afghanistan (N = 9, 16.4%) and Iraq (N = 6, 10.9%); for immigrant children main countries of birth were Belgium (N = 21, 33.9%), Spain (N = 8, 12.9%), The Netherlands (N = 6, 9.7%) and Morocco (N = 4, 6.5%). Of these immigrant pupils, 88.7% were first- or second-generation immigrants, several of them having parents who were outside of Belgium; 11.3% comprised third-generation immigrants: grandchildren of immigrants.

Table 1.

Demographic characteristics (%) of the participants.

	Total group	Refugees	Immigrants	χ² (p)
Gender	N = 120	N = 55	N = 62	3.005 (.083)
Male	53.3	56.4	40.3
Female	46.7	43.6	59.7
Age	N = 120	N = 55	N = 62	7.648 (.177)
8 years	23.3	25.5	21.0
9 years	27.5	32.7	22.6
10 years	21.7	20	22.6
11 years	23.3	14.5	32.3
12 years	4.2	7.3	1.5
Time in Belgium	N = 117	N = 52	N = 62	15.405 (.002)
<1 year	30.8	36.5	25.8
1–5 years	33.3	46.2	22.6
5–10 years	12.8	7.7	16.1
Born in Belgium	23.1	9.6	35.5
Legal status	N = 117	N = 53	N = 61	70.244 (.000)
Belgian nationality	29.9	15.1	42.6
Refugee status	27.4	60.4	0
Subsidiary protection	4.2	9.4	0
In asylum procedure	0.9	1.9	0
Undocumented	4.3	5.7	3.3
Other^a	33.3	7.5	54.1
Maternal education	N = 91	N = 34	N = 54	20.883 (<.001)
No schooling	29.7	58.8	13.0
Basic education	50.5	26.5	63.0
Higher education	19.8	14.7	24.0
Paternal education	N = 80	N = 33	N = 46	4.805 (.090)
No schooling	21.3	33.3	13.0
Basic education	56.2	45.5	63.0
Higher education	22.5	21.2	24.0

^a‘Other’ includes legal statuses not specified otherwise (e.g., a European work permit, residence permit without further specification).

Chi-Square tests of independence regarding demographic characteristics indicated a significant difference in terms of maternal education, with more mothers of refugee pupils having received no schooling (χ²(2) = 20.88; p < .001). In addition, there were significant differences between on migration-related variables, as could be expected: refugee and immigrant participants differed from each other in terms of legal status (χ²(5) = 70.24; p = .002); time in Belgium (χ²(3) = 15.41; p = .002); and country of birth (χ²(30) = 98.11; p = .001).

Measures and procedures

Strengths and difficulties questionnaire

The SDQ is a 25-item measure of emotional and behavioral problems in children (Goodman et al., 2000), with extensive application in culturally diverse study settings (including translations in over 40 languages) (Bourdon et al., 2005; Goodman, 2001). Five small subscales (emotional symptoms, conduct problems, hyperactivity, peer problems, and prosocial behavior); two broader subscales (externalizing and internalizing problems); and a total difficulties score can be computed. The self-report version is intended for use with 11- to 16-year-olds, yet has also been validated with younger children (Curvis et al., 2014). In our study we invited children, their parents and teachers to complete the SDQ. Except for the self-reported internalizing problems (α = .54), all three informants’ total difficulties scores, externalizing and internalizing subscale scores showed satisfactory to good internal consistency (α > .60; Taber, 2018). Several of the smaller subscales of the SDQ had Cronbach’s alphas below .60 and were not further studied, although the use of these smaller subscales is generally more recommended in high-risk samples (Goodman et al., 2010).

Youth self report – post-traumatic stress problems scale

The Youth Self Report (YSR) is part of the Achenbach System of Empirically Based Assessment (ASEBA). The ASEBA forms have been translated in over 100 languages (Achenbach & Rescorla, 2007). The YSR was developed for use with 11- to 18-year-olds, yet YSR scores for younger children (ages 7–10) show good psychometric properties (Ebesutani et al., 2011). In this study, only the 14 items of the post-traumatic stress problems scale of the Youth Self Report (PTSP; Achenbach & Rescorla, 2001; 2007) were included in the children’s self-report questionnaires. These items assess a broad range of post-trauma symptoms and trauma-specific behavioral problems, and result in a scale score ranging from 0 to 28, enabling the level of PTSP to be categorized as ‘normal’, ‘borderline’, or ‘clinical’ (α = .73).

Climate scale

Two 8-item subscales of the Climate Scale (CS; Donkers & Vermulst, 2014) were included in the self-report questionnaires. The Climate in the Class (CIC) subscale assesses children’s perceptions of classroom peer interactions. The Quality of Mutual Peer Relationships (QMP) subscale measures pupils’ satisfaction with the connectedness and cohesion among classroom peers. Standardization data were gathered among a large sample of pupils in The Netherlands, resulting in satisfactory norms for general primary schools (COTAN, 2014), though standardization for refugee and immigrant children is lacking. Studies have provided evidence for the validity of the subscales (Donkers & Vermulst, 2014). Cronbach’s alpha in our sample was .84 for the CIC subscale and .79 for QMP.

In the multi-informant data-collection (March 2018), teachers completed questionnaires independently; data collection with children and parents was organized in (class) group, assisted by the research team. Children and parents received additional support by means of pictograms, translation by qualified interpreters, translated questionnaires, and standardized clarifications by the researchers. Participants worked independently. Socially desirable responses were discouraged as much as possible by ensuring participants that their answers would be treated confidentially. Prior to the study, approval of the research design was obtained from the university’s Ethics Committee.

Analysis

Descriptive statistics were used to describe demographic characteristics and the outcome variables for the refugee, the immigrant and the total sample. Given that the assumption of normality was not met for several of the variables in our participant sample, non-parametric Mann-Whitney tests were performed to compare the means of both groups. In addition, the clinical significance of outcome measures was examined using categorizations and cut-off scores based on large population-based surveys. Finally, cross-tabulations and kappa statistics were used to investigate the level of agreement on the amount of total difficulties between the multiple informants that completed the SDQ (Tang et al., 2015).

Results

Mental health problems

The means and standard deviations of all dependent variables measuring mental health are presented in Table 2. We found no significant differences between refugee and immigrant participants on the different outcome variables, with exception for externalizing problems reported by parents: externalizing problems were higher for immigrant pupils (M = 5.92) than for refugee pupils (M = 4.49), U = 1239.5, p = .043.

Table 2.

Mental health problems in primary school refugee and immigrant children.

Scale	Total group	Refugees	Immigrants	U (p)
SDQ-SELF	N = 117	N = 53	N = 61
Total problems	11.66 (5.46)	11.11 (5.26)	12.13 (5.49)	1448 (.337)
Internalizing^a	6.50 (3.25)	6.28 (3.10)	6.64 (3.32)	1526 (.606)
Externalizing^b	5.16 (3.30)	4.83 (3.46)	5.49 (3.14)	1408 (.234)
SDQ-Parent	N = 115	N = 53	N = 60
Total problems	10.30 (6.21)	9.66 (6.43)	10.90 (6.04)	1357.5 (.180)
Internalizing	5.03 (3.55)	5.17 (3.62)	4.98 (3.55)	1622.5 (.851)
Externalizing	5.22 (3.73)	4.49 (3.41)	5.92 (3.77)	1239.5 (.043)
SDQ-Teacher	N = 117	N = 54	N = 60
Total problems	6.92 (5.10)	6.98 (4.98)	6.87 (5.36)	1683 (.720)
Internalizing	2.79 (2.88)	2.93 (2.83)	2.72 (2.96)	1711.5 (.598)
Externalizing	4.13 (3.80)	4.06 (3.74)	4.15 (3.94)	1601.5 (.916)
YSR	N = 117	N = 53	N = 61	1500 (.507)
PTSP	10.17 (4.99)	9.83 (4.82)	10.43 (4.97)	1500 (.507)

Mean (SD). Minimum and maximum scores: SDQ: total problems: 0–40, internalizing and externalizing problems: 0–20; YSR-PTSP: 0–28.

^aInternalizing subscale: emotional symptoms + peer problems subscales.

^bExternalizing subscale: hyperactivity + conduct problems subscales.

To facilitate the interpretation of the clinical significance of these descriptive statistics, we categorized participants’ scores using cut-off scores reported in large population-based surveys. For the SDQ, we employed the most recent fourfold classification to indicate whether values on the total problems scale were close to average, slightly raised, high, or very high (YouthinMind, 2016). For the PTSP subscale, we used the cut-off scores for borderline and clinical range as indicated in the ASEBA manual (Achenbach & Rescorla, 2007). Group 2 (standard) norms were used, given that normative data were lacking for almost all participants’ home societies as well as for the host country. According to these norms, a raw score of 12–14 for boys and 14–17 for girls is equivalent to the borderline range (T score of 64–69), whereas a raw score of 15 or higher for boys and 18 or higher for girls corresponds to a clinical level (T score ≥70). Boys and girls were categorized separately, subsequently percentages were calculated for boys and girls together. Table 3 summarizes this clinical categorization (%). Around 25% of participants score high or very high on emotional and behavioral difficulties (SDQ). Around 30% score in the borderline or clinical range on post-traumatic stress problems (YSR). Contrary to our hypothesis, immigrant participants also showed high rates of total difficulties and post-traumatic stress problems.

Table 3.

Clinical categorization (%) of SDQ and YSR scores.

	Total group	Refugees	Immigrants	X² (p)
SDQ-self total	N = 117	N = 53	N = 61	1.206 (.752)
Close to average	48.7	50.9	45.9
Slightly raised	25.7	24.5	27.9
High	12.8	15.2	11.5
Very high	12.8	9.4	14.7
SDQ-parent total	N = 115	N = 53	N = 60	4.325 (.228)
Close to average	61.7	69.8	55
Slightly raised	17.4	9.4	23.3
High	13.9	13.3	15
Very high	7	7.5	6.7
SDQ-teacher total	N = 117	N = 54	N = 60	3.628 (.305)
Close to average	79.5	75.9	81.7
Slightly raised	15.4	18.5	13.3
High	1.7	3.7	0
Very high	3.4	1.9	5
YSR	N = 117	N = 53	N = 61	1.225 (.542)
Normal	69.2	71.7	67.2
Borderline	17.1	18.9	16.4
Clinical	13.7	9.4	16.4

Table 4 reports information on the degree of agreement between the three pairs of raters of SDQ total difficulties (self-parent; self-teacher; parent-teacher). Cohen’s kappa statistics were calculated to determine the proportion of agreement between the three pairs of raters over and above the agreement expected by chance (Tang et al., 2015). There was fair agreement between children and parents (κ = .22; p = .016); slight agreement between parents and teachers (κ = .13; p = .08); and slight agreement between children and teachers (κ = .09; p =. 156) (Landis & Koch, 1977). In this latter dyad, 22.8% of children reporting high or very high difficulties were not detected by their teacher.

Table 4.

Agreement between informants on SDQ-Total difficulties in the child (%).

	Parent		Teacher
	Normal	Abnormal	Normal	Abnormal
Self
Normal	61.9	11.5	71.9	2.6
Abnormal	16.8	9.7	22.8	2.6
Parent
Normal			75.9	2.7
Abnormal			18.8	2.7

Normal: average or slightly raised. Abnormal: high or very high.

Social classroom relationships

Means and standard deviations of the CS subscales are presented in Table 5. Mann-Whitney tests revealed no significant differences between the refugee and immigrant subgroups.

Table 5.

Social classroom relationships as reported by primary school refugee and immigrant children.

	Total group	Refugees	Immigrants	U (p)
Climate in the class	N = 117	N = 53	N = 61
Raw score	2.76 (.74)	2.75 (.77)	2.77 (.70)	1600 (.925)
Transformed score	2.92 (1.57)	2.98 (1.65)	2.85 (1.50)	1648 (.855)
Quality of mutual peer relationships	N = 117	N = 53	N = 61
Raw score	3.14 (.64)	3.10 (.68)	3.17 (.62)	1536.5 (.649)
Transformed score	3.68 (1.67)	3.57 (1.76)	3.77 (1.62)	1504.5 (.514)

Mean (SD). Minimum and maximum scores: raw score: 1–4; transformed score: 1–7, with ‘1’ indicating very negative perceptions of class climate and quality of peer relations and ‘7’ very positive perceptions (1 = very low; 2 = low; 3 = below average; 4 = average; 5 = above average; 6 = high; 7 = very high).

We compared scores to the available norm group of general primary education pupils in The Netherlands. The frequency distribution of the subscale scores obtained within this norm group was transformed into a uniform septile distribution, with consecutive categories containing 5, 10, 20, 30, 20, 10, and 5% of the scores. The average transformed scores indicated that the total participant group as well as the two subgroups perceive the quality of mutual peer relationships between ‘below average’ and ‘average’. The climate in the class is rated lower, with norms indicating ‘low’ to ‘below average’ scores. Table 6 shows the categorization (%) of scores according to the septile distribution.

Table 6.

Categorization (%) of climate scale scores.

	Total group	Refugees	Immigrants
Climate in the class	N = 117	N = 53	N = 61
Very low	22.2	18.9	24.6
Low	19.7	24.5	16.4
Below average	25.6	24.5	27.9
Average	19.7	20.8	18
Above average	6.8	1.9	9.8
High	0.9	1.9	0
Very high	5.1	7.5	3.3
Quality of mutual peer relationships	N = 117	N = 53	N = 61
Very low	11.1	13.2	9.8
Low	12.8	18.9	8.2
Below average	20.5	13.2	24.6
Average	32.5	32.1	34.4
Above average	8.5	7.5	8.2
High	5.1	5.7	5
Very high	9.5	9.4	9.8

In Table 7, correlations between measures are reported. It is noteworthy that CIC scores correlate negatively with self- and parent-reported emotional and behavioral difficulties (r = −.354, p < .01 and r = −.191, p < .05 respectively), as well as with PTSP (r = −.265, p < .01).

Table 7.

Pearson correlations between mental health outcomes and social classroom relationships.

	SDQ-Self	SDQ-Parent	SDQ-Teacher	PTSP	Class climate	Peer relations
SDQ-Self	1	.293 (.002)**	.208 (.026)*	.701 (.000)**	−.354 (.000)**	−.400 (.000)**
SDQ-Parent		1	.322 (.001)**	.253 (.007)**	−.191 (.043)*	−.177 (.060)
SDQ-Teacher			1	.099 (.294)	.096 (.312)	−.080 (.396)
YSR				1	−.265 (.004)**	−.174 (.061)
Climate in the class					1	.362 (.000)**
Quality of mutual peer relationships						1

p-values are indicated between brackets: **Correlation is significant at the .01 level (2-tailed); *Correlation is significant at the .05 level (2-tailed).

Discussion

Through a multi-informant assessment, this study explored mental health profiles and classroom relationships in primary school-aged refugee and immigrant children.

This study shows self-reported mental health problems that are elevated as compared to those documented in population-based surveys: on emotional and behavioral problems, around 25% of participants scored high or very high, while this percentage in European community samples generally lies around 10% (e.g., Wright et al., 2020). On PTSP, around 30% of the total participant sample scored in the borderline or clinical range, while prevalence of PTSD has been estimated at <1–10% in the general youth population (Mollica et al., 2004). Comparing our study results to other studies using this YSR subscale, the percentage of participants scoring in the borderline or clinical range corresponds to that found in poly-victimized adolescents (Kirchner et al., 2014). Scores for emotional and behavioral problems, nor for PTSP, statistically differed for refugee versus immigrant participants, with exception of immigrants’ parents reporting more externalizing problems than refugees’ parents.

For refugee children, 28.3% scored borderline or clinical on PTSP, which lies within the range of estimated prevalence rates of PTSD for this group, situated between 19 and 54% in previous studies (Kien et al., 2019). The 9.4% of refugee participants scoring ‘very high’ on emotional and behavioral difficulties echoes findings of Dutch study with asylum-seeking children (9%; Wiegersma et al., 2011). Immigrant participants, almost 90% of which were first or second generation immigrants, reported similar rates of emotional and behavioral difficulties as refugee participants. Refugee and immigrant children share resettlement stressors of social isolation, elevated poverty risks, discrimination, and lack of belonging, with previous studies demonstrating how these risk factors increase psychological problems among immigrant children (Ceri et al., 2017). Remarkably, PTSP-rates in immigrant participants were high, and indicate the importance of further understanding immigrant children’s trauma exposure and its mental health impact. Indeed, with a present-day lack of quantitative studies on posttraumatic stress in non-refugee immigrant children, our study points to the need of adopting sensitivity to how traumatization may impact these children’s psychosocial functioning. With few studies pointing to the role of community violence (Jaycox et al., 2002), domestic violence (e.g., Gray et al., 2014), and traumatic stressors during migration, future research is needed to support an in-depth understanding of trauma exposure in immigrant children.

Whereas around 25% of participant children report a high or very high prevalence of emotional and behavioral problems, teachers make a less alarming assessment. They locate 5% of participants within the range of (very) high symptomatic functioning. Cross-informant comparisons indicate that an important group (22.8%) of children reporting (very) high difficulties is not detected by their teacher. A comparison of mean scores furthermore indicates that teachers identify different symptoms: while children report more internalizing than externalizing problems, teachers observe more externalizing symptomatology. This low correspondence is in line with previous research that indicated low teacher-child correspondence, with teachers perceiving a lot less internalizing problems than refugee children themselves (Goldin et al., 2008). These findings underline the importance of focusing on the advancement of teachers’ expertise in the early identification of children with internalizing difficulties (Weiss et al., 2019).

Another remarkable finding of our study was the relatively conflictual and unsafe appraisal of classroom atmosphere by the majority (70%) of participants. The percentages of very low and low ratings are much higher in our sample than the ones found in population-based standardization research, where only 35% of children in the norm group rated the climate in the class as very low to below average (Donkers & Vermulst, 2014). In our participants, perceived unsafety in the classroom furthermore related to higher post-trauma symptomatology. PTSD is, among other symptoms, characterized by increased arousal (e.g., hypervigilance, irritability) and distrust (APA, 2013). This might result in difficult peer relationships, negative perceptions of classroom peer interactions, or withdrawal, and potentially intersect with bereavement or missing a (sometimes idealized) home country. On the other hand, an unsafe classroom context, characterized by conflicts and bullying, might increase PTSD symptomatology in traumatized children, potentially also compounding previous migration-related traumatic experiences (Kira et al., 2014). Given that previous research has associated a perceived sense of safety at school with low risk of PTSD (Geltman et al., 2005), it seems crucial to make classroom safety and positive peer interactions central foci in school-based interventions aiming to improve young refugees’ and immigrants’ well-being. These perspectives highlight also the importance of supporting teachers in sensitively identifying and regulating traumatic distress in their pupils.

Limitations and future directions

Findings are based on a relatively small sample size, associated with difficulties in recruitment of refugee families. While relatively small sample sizes are generally accepted in this field due to the uniqueness of the study population and the urgency of the public health issue under study (e.g., Fête et al., 2019), studies with larger samples remain pivotal.

Our study was implemented in three schools with a high concentration of immigrants in school composition. While schools frequented by refugee and immigrant children are often characterized by multi-ethnic school composition with a lower socio-economic profile in an urban setting (Pulinx et al., 2017), the high number of refugee and immigrant children in participating schools may limit the generalizability of our study’s findings to similarly composed school contexts. Future research on refugee and immigrant children’s mental health might furthermore consider ethnic school composition in itself, as scholars documented the role of school composition in pupils’ self-esteem and the quality of the interethnic peer relationships (e.g., Agirdag et al., 2012).

Concerning assessment instruments, both SDQ and YSR were originally developed for use with youth aged 11 and older, while a substantial part of our participant sample was younger. Although reliability and validity of these scales for younger children has been studied (Curvis et al., 2014; Ebesutani et al., 2011), the diverse cultural, linguistic, and migratory background of the participants might have further complicated assessment (Stolk et al., 2017). Also, while overall the internal consistency of measures proved satisfactory, the lower internal consistency of self-reported internalizing problems (SDQ) necessitates careful interpretation of our findings.

Lastly, our aim was to describe mental health in populations of refugee and immigrant children. While differences were found between both groups, this does not necessarily mean that these differences can be attributed to children’s refugee status. Indeed, we found that both groups differed in various aspects (e.g., maternal education) and it is possible that differences in mental health are due to these (or other) aspects instead. Unfortunately, our relatively small sample size did not allow for an in-depth comparison of refugees’ and immigrants’ mental health profiles, taking into account potentially confounding factors such as gender, time in resettlement, socio-economic profile, and maternal educational status. This is especially true because some potential confounding variables are categorical with few observations per category and because of missing values. With our study raising hypotheses on the potential role of different migration-related and socio-economic factors in both refugee and immigrant subsamples, future large-scale studies may develop a better understanding of the interplay between these variables and refugee and immigrant child mental health (Dogra et al., 2012).

Conclusion and implications

This study reports findings with important public health implications: it offers insight into the vulnerable mental health profiles of refugee and immigrant children in multi-ethnic classrooms in an urban Flemish sample. High levels of emotional and behavioral difficulties and post-trauma functioning in immigrant pupils indicate that not only refugee, but also immigrant children form a population at risk for traumatization. Further research is needed on forms of past and current trauma exposure. The study also highlights high degrees of stress and conflict experienced in the classroom by children, indicating the potential role of school-based intervention in promoting safety and well-being for an at-risk group.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Fund for Scientific Research Flanders (FWO – Fonds voor Wetenschappelijk Onderzoek) under Grant number G0F5117N (FWO-Rode Neuzen)

ORCID iDs

Ruth Kevers

Caroline Spaas

Wim Van Den Noortgate

Author biographies

Ruth Kevers, PhD, works as a family therapist in the child psychiatry daycare unit PASO of the University Psychiatry Hospital (UPC KU Leuven & PraxisP). She is a clinical member in the Transcultural Trauma Care for Refugees team at the Clinical Centre PraxisP of the Faculty of Psychology and Educational Sciences, KU Leuven.

Caroline Spaas is a clinical child psychologist, family therapist and doctoral researcher at KU Leuven, Faculty of Psychology and Educational Sciences. Her research focuses on relational processes of post-trauma reconstruction in community-based care for refugee children, adolescents and their families.

Hilde Colpin is Full Professor at KU Leuven, Faculty of Psychology and Educational Sciences. She conducts longitudinal and intervention research regarding the effect of the social context on children’s and adolescents’ social-emotional and academic development.

Wim Van Den Noortgate is full professor at the Faculty of Psychology and Educational Sciences and at itec, an imec research group at KU Leuven. His major research interest is in statistical modelling of educational data, especially in the field of meta-analysis and learning analytics.

Sofie de Smet is a postdoctoral researcher at KU Leuven, Faculty of Psychology and Educational Sciences and at Ghent University, Faculty of Arts and Philosophy. Her current interdisciplinary postdoctoral research focuses on the development of community-based refugee posttrauma care in the field of transcultural psychology and participatory theatre in western host societies, including an analysis of the role of trauma narration in community-based mental health interventions.

Ilse Derluyn is Full Professor at Ghent University, Belgium. Her major research interests include the wellbeing of refugee children, war-affected young people and victims of trafficking. Ilse is ERC-grant holder of the research project Child Move and coordinates the H2020-project Refugees WellSchool. She is director of the Centre for the Social Study of Migration and Refugees (CESSMIR) at Ghent University and co-director of the Centre for Children in Vulnerable Situations.

Lucia De Haene is Associate Professor at the Faculty of Psychology & Educational Sciences, KU Leuven, where she is involved in research on refugee family relationships and mental health intervention with refugee families and communities. As family therapist, she coordinates her team in Transcultural Trauma Care for Refugees at the Faculty Clinical Centre PraxisP, and leads the development of the child psychiatry daycare unit PASO of the University Psychiatry Hospital (UPC KU Leuven & PraxisP).

References

Achenbach

T. M.

Rescorla

L. A.

(2001). Manual for the ASEBA school-age forms & profiles: University of Vermont.

Achenbach

T. M.

Rescorla

L. A.

(2007). Multicultural supplement of the manual for the ASEBA school-age forms and profiles: University of Vermont.

Agirdag

Van Houtte

Van Avermaet

(2012). Ethnic school segregation and self-esteem: The role of teacher-pupil relationships. Urban Education, 47(6), 1135–1159. https://doi.org/10.1177/0042085912452154

American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed.). American Psychiatric Association.

Betancourt

T. S.

Newnham

E. A.

Birman

Lee

Ellis

B. H.

Layne

C. M.

(2017). Comparing trauma exposure, mental health needs, and service utilization across clinical samples of refugee, immigrant, and US‐origin children. Journal of Traumatic Stress, 30(3), 209–218. https://doi.org/10.1002/jts.22186

Bourdon

Goodman

Rae

Simpson

Koretz

(2005). The strengths and difficulties questionnaire: US normative data and psychometric properties. Journal of the American Academy of Child and Adolescent Psychiatry, 44(6), 557–564. https://doi.org/10.1097/01.chi.0000159157.57075.c8

Ceri

Özlü-Erkilic

Özer

Ü.

Kadak

Winkler

Dogangün

Akkaya-Kalayci

(2017). Mental health problems of second generation children and adolescents with migration background. International Journal of Psychiatry in Clinical Practice, 21(2), 142–147. https://doi.org/10.1080/13651501.2017.1291820

COTAN (2014). Beoordeling van de Klimaatschaal [Evaluation of the climate scale]. https://www.cotandocumentatie.nl/beoordelingen/b/14784/klimaatschaal/

Curvis

McNulty

Qualter

(2014). The validation of the self-report strengths and difficulties questionnaire for use by 6- to 10-year-old children in the UK. British Journal of Clinical Psychology, 53(1), 131–137. https://doi.org/10.1111/bjc.12025

10.

Derluyn

Broekaert

Schuyten

(2008). Emotional and behavioural problems in migrant adolescents in Belgium. European Child & Adolescent Psychiatry, 17(1), 54–62. https://doi.org/10.1007/s00787-007-0636-x

11.

De Haene

Grietens

Verschueren

(2010a). Holding harm: Narrative methods in mental health research on refugee trauma. Qualitative Health Research, 20(12), 1664–1676. https://doi.org/10.1177%2F1049732310376521

12.

De Haene

Grietens

Verschueren

(2010b). Adult attachment in the context of refugee traumatisation: the impact of organized violence and forced separation on parental states of mind regarding attachment. Attachment & Human Development, 12(3), 249–264. https://doi.org/10.1080/14616731003759732

13.

de Smet

Rousseau

Stalpaert

De Haene

(2020). I look at you and see you looking at me”: Role boundaries in a dynamic research relationship in qualitative health research with refugees. Qualitative Health Research, 30(7), 1083–1100. https://doi.org/10.1177/1049732320910411

14.

Dogra

Singh

S. P

Svirydzenka

Vostanis

(2012). Mental health problems in children and young people from minority ethnic groups: the need for targeted research. The British Journal of Psychiatry, 200(4), 265–267. https://doi.org/10.1192/bjp.bp.111.100982

15.

Donkers

Vermulst

A. A.

(2014). Klimaatschaal: Handleiding [climate scale: Manual]: Orthopraktijk Ad Donkers & Ad Vermulst.

16.

Due

Riggs

D. W.

Augoustinos

(2016). Experiences of school belonging for young children with refugee backgrounds. The Educational and Developmental Psychologist, 33(1), 33–53. https://doi.org/10.1017/edp.2016.9

17.

Ebesutani

Bernstein

Martinez

J. I.

Chorpita

B. F.

Weisz

J. R.

(2011). The youth self report: Applicability and validity across younger and older youths. Journal of Clinical Child & Adolescent Psychology, 40(2), 338–346. https://doi.org/10.1080/15374416.2011.546041

18.

Fazel

(2015). A moment of change: Facilitating refugee children’s mental health in UK schools. International Journal of Educational Development, 41(2015), 255–261. https://dx-doi-org.web.bisu.edu.cn/10.1016/j.ijedudev.2014.12.006

19.

Fête

Aho

Benoit

Cloos

Ridde

(2019). Barriers and recruitment strategies for precarious status migrants in Montreal, Canada. BMC Medical Research Methodology, 19(41), 1–14. https://doi.org/10.1186/s12874-019-0683-2

20.

Geltman

P. L.

Grant-Knight

Mehta

S. D.

Lloyd-Travaglini

Lustig

Landgraf

J. M.

Wise

P. H.

(2005). The “lost boys of Sudan”: Functional and behavioral health of unaccompanied refugee minors resettled in the United States. Archives of Pediatrics & Adolescent Medicine, 159(6), 585–591. https://doi.org/10.1001/archpedi.159.6.585

21.

Gharaei

Thijs

Verkuyten

(2019). Ethnic identity in diverse schools: Preadolescents’ private regard and introjection in relation to classroom norms and composition. Journal of Youth and Adolescence, 48(1), 132–144. https://doi.org/10.1007/s10964-018-0881-y

22.

Goldin

Hägglöf

Levin

Persson

L.A.

(2008). Mental health of Bosnian refugee children: A comparison of clinician appraisal with parent, child and teacher reports. Nordic Journal of Psychiatry, 62(3), 204–216. https://doi.org/10.1080/08039480801983604

23.

Goodman

Lamping

D. L.

Ploubidis

G. B.

(2010). When to use broader internalising and externalising subscales instead of the hypothesised five subscales on the strengths and difficulties questionnaire (SDQ): Data from British parents, teachers and children. Journal of Abnormal Child Psychology, 38(8), 1179–1191. https://doi.org/10.1007/s10802-010-9434-x

24.

Goodman

(2001). Psychometric properties of the strengths and difficulties questionnaire (SDQ). Journal of the American Academy of Child and Adolescent Psychiatry, 40(11), 1337–1345. https://doi.org/10.1097/00004583-200111000-00015

25.

Goodman

Ford

Simmons

Gatward

Meltzer

(2000). Using the strengths and difficulties questionnaire (SDQ) to screen for child psychiatric disorders in a community sample. British Journal of Psychiatry, 177(6), 534–539. https://doi.org/10.1192/bjp.177.6.534

26.

Gray

Easteal

Bartels

(2014). Immigrant women and family violence: Will the new exceptions help or hinder victims? Alternative Law Journal, 39(3), 167–171. https://doi.org/10.13140/2.1.2487.7766

27.

Hodes

Vostanis

(2019). Practitioner review: Mental health problems of refugee children and adolescents and their management. Journal of Child Psychology and Psychiatry, 60(7), 71–731. https://doi.org/10.1111/jcpp.13002

28.

Jaycox

L. H.

Stein

B. D.

Kataoka

S. H.

Wong

Fink

Escudero

P. I. A.

Zaragoza

(2002). Violence exposure, posttraumatic stress disorder, and depressive symptoms among recent immigrant schoolchildren. Journal of the American Academy of Child and Adolescent Psychiatry, 41(9), 1104–1110. https://doi.org/10.1097/00004583-200209000-00011

29.

Kien

Sommer

Faustmann

Gibson

Schneider

Krczal

Jank

Klerings

Szelag

Kerschner

Gartlehner

Brattström

(2019). Prevalence of mental disorders in young refugees and asylum seekers in European countries: A systematic review. European Child & Adolescent Psychiatry, 28, 1295–1310. https://doi.org/10.1007/s00787-018-1215-z

30.

Kira

I. A.

Lewandowski

Ashby

J. S.

Somers

Chiodo

Odenat

(2014). Does bullying victimization suppress IQ? The effects of bullying victimization on IQ in Iraqi and African American adolescents: A traumatology perspective. Journal of Aggression, Maltreatment & Trauma, 23(5), 431–453. https://doi.org/10.1080/10926771.2014.904463

31.

Kirchner

Forns

Soler

Planellas

(2014). Post-traumatic stress problems among poly-victimized Spanish youth: Time effect of past vs. recent interpersonal victimizations. Child Abuse & Neglect, 38(8), 1303–1312. https://doi.org/10.1016/j.chiabu.2014.02.014

32.

Kouider

E. B.

Koglin

Petermann

(2014). Emotional and behavioral problems in migrant children and adolescents in Europe: A systematic review. European Child & Adolescent Psychiatry, 23(6), 373–391. https://doi.org/10.1007/s00787-013-0485-8

33.

Landis

J. R.

Koch

G. G.

(1977). The measurement of observer agreement for categorical data. Biometrics, 33(1), 159–174. https://doi.org/10.2307/2529310

34.

Mollica

R.F.

Cardozo

B.L.

Osofsky

H.J.

Raphael

Ager

Salama

(2004). Mental health in complex emergencies. Lancet, 364(9450), 2058–2067. https://doi.org/10.1016/S0140-6736(04)17519-3

35.

Pulinx

Kemper

Verswijvel

De Bruyne

Vanthournout

Baeten

Keygnaert

(2017). Cartografie en analyse van het onthaalonderwijs voor anderstalige nieuwkomers en OKAN-leerlingen: het onthaalonderwijs voor anderstalige nieuwkomers en OKAN-leerlingen in kaart brengen en kritisch analyseren. Leuven: Gent/Leuven/Antwerpen. https://data-onderwijs.vlaanderen.be/documenten/bestand.ashx?nr=9679

36.

Rousseau

Singh

Lacroix

Bagilishya

Measham

(2004). Creative expression workshops for immigrant and refugee children. Journal of the American Academy of Child and Adolescent Psychiatry, 43(2), 235–238. https://doi.org/10.1097/00004583-200402000-00021

37.

Spaas

Verelst

Devlieger

Aalto

Andersen

A. J.

Durbeej

, et al. (2022) (In press). Mental health of refugee and non-refugee migrant young people in European Secondary Education: the role of family separation, daily material stress and perceived discrimination in resettlement. Journal of Youth and Adolescence, 51, 848–870. https://doi.org/10.1007/s10964-021-01515-y

38.

Stolk

Kaplan

Szwarc

(2017). Review of the strengths and difficulties questionnaire translated into languages spoken by children and adolescents of refugee background. International Journal of Methods in Psychiatric Research, 26(4), Article e1568. https://doi.org/10.1002/mpr.1568

39.

Taber

K.S.

(2018). The use of Cronbach’s alpha when developing and reporting research instruments in science education. Research in Science Education, 48(6), 1273–1296. https://doi.org/10.1007/s11165-016-9602-2

40.

Tang

Zhang

(2015). Kappa coefficient: A popular measure of rater agreement. Shanghai Archives of Psychiatry, 27(1), 62–67. https://doi.org/10.11919/j.issn.1002-0829.215010

41.

Vervliet

Meyer Demott

M. A.

Jakobsen

Broekaert

Heir

Derluyn

(2014). The mental health of unaccompanied refugee minors on arrival in the host country. Scandinavian Journal of Psychology, 55(1), 33–37. https://doi.org/10.1111/sjop.12094

42.

Weiss

Syring

Kiel

(2019). Challenges for the school in a migration society – a critical incidents analysis. Intercultural Education, 30(2), 196–213. https://doi-org.kuleuven.e-bronnen.be/10.1080/14675986.2019.1538089

43.

Wiegersma

P. A.

Stellinga-Boelen

A. A.

Reijneveld

S. A.

(2011). Psychosocial problems in asylum seekers’ children: The parent, child, and teacher perspective using the Strengths and Difficulties Questionnaire. The Journal of Nervous and Mental Disease, 199(2), 85–90. https://doi.org/10.1097/NMD.0b013e31820446d2

44.

Wright

Garside

Allgar

Hodkinson

Thorpe

(2020). A large population-based study of the mental health and wellbeing of children and young people in the North of England. Clinical Child Psychology and Psychiatry, 25(4), 877–890. https://doi.org/10.1177/1359104520925873

45.

YouthinMind (2016). Scoring the strengths and difficulties questionnaire fora ge 4-17 or 18+. August 23, 2019 https://www.sdqinfo.com/py/sdqinfo/c0.py