Emotional and behavioural problems
in migrant and non-migrant adolescents in Belgium
Ilse Derluyn, MEduc
Eric Broekaert, PhD
Gilberte Schuyten, PhD
Refugee and migrant adolescents may be at increased risk for the development of emotional
and behavioural problems.
This study aims to investigate the emotional and behavioural problems of newly arrived
adolescent migrants in Belgium compared to Belgian peers.
1249 migrant adolescents and 602 Belgian adolescents filled in four questionnaires on
emotional and behavioural problems, traumatic experiences and post-traumatic stress.
Migrants have more peer problems and avoidance symptoms than non-migrants; on the
contrary, non-migrant adolescents have more anxiety symptoms, externalising problems and
hyperactivity. Migrants experienced more traumatic events than non-migrants. Important
influencing factors in the development of emotional and behavioural problems were the
number of traumatic events, gender and living situation.
Migrants and non-migrant don’t differ much in the development of emotional and behavioural
symptoms. Risk groups for the development of emotional problems and post-traumatic stress
are migrants who experienced a lot of traumatic events, girls and unaccompanied migrant
Declaration of interest
Funding detailed in Acknowledgements.
With the increasing number of migrants and refugees worldwide, the effect of migration and
acculturation on psychological development has become very important (Stevens, Pels,
Bengi-Arslan, et al, 2003), especially for children and adolescents. Migrating children may
face traumatic events before or during the flight (Berman, 2001), have to cope with different
losses and separations (Perez Foster, 2001), and are forced to re-establish their life in an
unfamiliar environment. Therefore, this migration process can result in the development of
emotional and behavioural problems (Lavik, Hauff, Skrondal, et al, 1996), such as post-
traumatic stress, depression, anxiety, fear of recurrence, guilt, separation fears, grief,
withdrawal, eating and sleeping problems, identity confusion, and delinquent behaviour
(Allwood, Bell-Dolan & Husain, 2002). Recent research, however, has produced
contradictory results about emotional and behavioural problems in migrant adolescents, with
some studies suggesting they are at increased risk, and others demonstrating no differences or
a lower risk among migrants than among the native-born (Davies & McKelvey, 1998; Sam,
1994). Moreover, in Belgium, research about the psychological problems of migrant
adolescents is scarce. Therefore, this study aims to investigate the emotional and behavioural
problems of newly arrived adolescent migrants in Belgium, compared to native-born peers.
Study setting and design
The study was conducted in intensive language classes for newly arrived, 11- to 18-years
olds, non-Dutch speaking migrant adolescents in Flanders (Belgium). From the 42 schools
organizing these classes, 34 were randomly selected to participate. After a classical
description of the study to the participants, their written informed consent was obtained and
they completed, anonymous, four self-report questionnaires under supervision of at least two
research assistants (educated to master’s level) during a 60-90 minute classroom period. As
all questionnaires were translated in different languages, the adolescents were given, as much
as possible, the questionnaires in the own mother tongue. There were 1294 migrant
adolescents who agreed to fill in the questionnaires, although not all participants completed
all items. As there is a continuous inflow of new adolescent migrants during the school year
from September until June, this is between 65% and 97% of the population in intensive
language classes for 11- to 18-years olds in Flanders. In further analyses, 45 migrants born in
western countries were excluded (43 from Western Europe, 1 from North America and 1 from
Japan), just as three migrants older than 21 years of age.
We included a control group of Belgian adolescents in the study: out of the five Flemish
provinces, 17 schools for 11- to 18-years old adolescents were randomly selected to
participate. We aimed a well-balanced sample concerning the ages of the adolescents and
their study programme. In total, 617 Belgian adolescents participated; 15 adolescents with
another than Belgian nationality were left out in further analyses. The research procedure was
similar to the one of the migrant adolescents.
Ethics approval for the study was given by the Ethics Committee of the Faculty of Psychology
and Educational Sciences, Ghent University.
Demographic and social characteristics, such as gender, age, nationality, time in Belgium and
living situation, were measured using a series of closed questions.
The Hopkins Symptom Checklist-37 for Adolescents (HSCL-37A), an adaptation of the
Hopkins Symptom Checklist-25 (HSCL-25) (Derogatis, Lipman, Rickels, et al, 1974; Parloff,
Kelman & Frank, 1954) was used. The HSCL-25 is a self-report screening tool used to
diagnose elevated symptoms of anxiety (ten items) and depression (fifteen items). Bean,
Spinhoven, Eurelings-Bontekoe, et al (in press) adapted the HSCL-25 for use with migrant
adolescents into the HSCL-37A: to overcome language problems, item questions were
simplified and shortened, and the questionnaire was translated into nineteen different
languages. In addition, a subscale of twelve items questioning different externalising
symptoms of substance use and aggressive behaviour was added to the original questionnaire.
All items must be answered on a Likert-scale, ranging from 1 (little) to 4 (very much). The
HSCL-25 has demonstrated its usefulness as screening tool for depression and anxiety in
various cross-cultural settings and with different refugee populations (e.g., Kleijn, Hovens &
Rodenburg, 2001; Lavik, Laake, Hauff, et al, 1999; Sandanger, Moum, Ingebrigtsen, et al,
The self-report version for 11- to 16-year olds of the Strengths and Difficulties Questionnaire
(SDQ) (Goodman, Meltzer & Bailey, 1998) is a behavioural screening questionnaire that asks
about 25 attributes. The twenty-five items are divided between five subscales with five items
each, generating scores for emotional symptoms, conduct problems, inattention-hyperactivity,
peer problems and pro-social behaviour. All scales, but the last, are summed to generate a
total difficulties score. Respondents use a 3-point Likert scale to indicate how far each
attribute applies. In this study, the extended version of the SDQ was used, which assesses also
the impact of symptoms in terms of resultant distress, social impairment or burden for others
(Goodman, 1999). This questionnaire is available in more than twenty languages. Research
shows that the SDQ has a high reliability and validity (Goodman, 2001), and is able to
discriminate between children drawn from high- and low-risk samples and to screen for child
psychiatric disorders, also in non-western populations (Goodman, Ford, Simmons, et al, 2000;
Mullick & Goodman, 2001).
To investigate the traumatic events the adolescents experienced, the Stressful Life Events
(SLE) (Bean, 2000) was used. Participants are directed to indicate whether or not they
experienced twelve different kinds of stressful events, such as war, natural disaster, separation
from family, physical or sexual abuse. The SLE is available in nineteen languages.
The Self Inventory Checklist for Adolescents (ZIL-A) is an adaptation of the Self Inventory
Checklist (ZIL) (Hovens, Bramsen & van der Ploeg, 2000), a self-report questionnaire
looking for the level of post-traumatic stress symptoms. The twenty-two item scale, parallel to
the DSM-IV criteria for post-traumatic stress disorder (PTSD), can be divided into three
subscales: intrusion (six items), avoidance (nine items) and hyper-arousal (seven items).
Participants are directed to answer the items according to the events endorsed on the SLE, and
to score the items on a Likert scale, ranging from 1 (not at all) to 4 (always). The
questionnaire was adapted into the ZIL-A (Bean, Spinhoven, Eurelings-Bontekoe, et al, in
press) to be better understandable for migrant adolescents: the items were simplified,
shortened, and translated into nineteen languages.
Descriptive statistics were used to describe main social and demographic characteristics of the
migrant and non-migrant group. Multivariate analysis of covariance (MANCOVA, SPSS
version 11.0 for Windows) was carried out to examine the effects of group (migrants and non-
migrants), gender, living situation (both parents, one parent or other), age and total number of
traumatic events (covariate) on the subscales of the HSCL-37A, the SDQ and the ZIL-A. All
second order interactions between the factors were included in the model.
Binary logistic regression analyses were executed to look for the main influencing factors on
the behavioural and emotional problems in the migrant group. Independent variables in these
analyses were age, gender, living situation (both parents, father, mother, alone, and other),
and the total number of traumatic experiences. As dependent variables, the subscales of the
different questionnaires with following thresholds were used: for the HSCL-37A, we used the
recommended threshold of 1.75 for the subscales anxiety, depression and internalising
problems (Mollica, Wyshak, de Marneffe, et al, 1986); for the subscale externalising
problems, we could not apply this threshold as this subscale is not yet validated. For the SDQ,
we used the thresholds as recommended by Goodman (1997); the borderline and clinical
groups were put together. For the ZIL-A, a total score of greater than 52 is judged to indicate
a PTSD full-symptom diagnosis (Hovens, Bramsen & van der Ploeg, 2000).
Finally, binary logistic regression was used to examine the relationship between the type of
trauma exposure (SLE-items as independent variables) and the risk for PTSD (a PTSD full
symptom diagnosis as dependent variable) for both groups of migrants and non-migrants.
Quality of the subscales
Reliability analyses, executed on all subscales of the HSCL-37A, the SDQ and the ZIL-A,
revealed good alpha values (greater than 0.70), except for the subscales conduct problems
(0.62), hyperactivity (0.69), and peer problems (0.48). The assessment of the one-factor
model fit (AMOS 5) is given with the RMSEA goodness-of-fit measure. Following Browne &
Cudeck (1993), the fit of the one-factor model for the subscales externalising problems (0.11)
and hyperactivity (0.11) were poor, for avoidance (0.10) and hyper-arousal (0.10) mediocre,
and for the subscales anxiety (0.06), depression (0.04), emotional problems (0.03), conduct
problems (0.04), peer problems (0.06), pro-social behaviour (0.04) and intrusion (0.06) good.
Thus, although the rather low reliability values for the subscales conduct and peer problems,
these subscales show a good fit for the one-factor model.
Characteristics of the sample
Insert table 1
Some demographic characteristics of both groups are given in table 1. The Belgian
adolescents are somewhat older than the migrants (t=-10.85, d.f.=1066, P<0.001). 10% of the
migrants live in Belgium without parents, and almost 3% with only a brother or sister. Less
migrant (84.9%) than non-migrant adolescents (97.5%; ²=169.9, d.f.=8, P<0.001) live with
one or both parents.
The migrant adolescents come from 93 different countries; the most frequent countries of
origin are Morocco (179, 14.4%), Ghana (135, 10.8%), Turkey (118, 9.5%), and Angola (40,
3.2%). 508 (41%) migrants come from Africa, 433 (34.9%) from Asia, 240 (19.4%) from
Eastern Europe, and 58 (4.7%) from South and Central America. 29 (2.6%) migrants arrived
in Belgium less than one month ago, 414 (37.6%) one to six months ago, 407 (37%) are in
Belgium for six months up to one year, and 251 (22.8%) for more than one year.
178 (14.5%) migrants live in an asylum center; 103 (58%) of them are unaccompanied
refugee minors. Ten migrant adolescents (0.8%) live in another center for minors, and fifteen
(1.2%) migrants and one non-migrant (0.2%) live in a foster family.
Insert table 2
The traumatic events the migrant and non-migrant adolescents experienced during their life
time, as indicated on the Stressful Life Events, are given in table 2. Migrants experienced an
average of 3.64 (s.d. 2.74) traumatic events, which is significant more (t=8.493; d.f.=1549;
P<0.001, 95% CI 0.75-1.19) than non-migrant adolescents (mean 2.67; s.d. 2.02). Most
traumatic events were experienced more by migrant adolescents; only the life event
‘experienced a stressful life event where he/she thought that someone else was in danger’ was
experienced more by non-migrants (P<0.03). For the events ‘the death of a loved one’,
‘sexual abuse’ and ‘experienced a stressful life event where he/she thought I am in danger’,
no significant differences between the groups were found. The death of a loved one was the
most experienced traumatic event: more than 56% of the migrants and 60% non-migrants
were confronted with it. More than 44% of the migrants and almost one third of the non-
migrants experienced important changes in their family life in the last year (OR 1.64); 23.5%
migrants were once in their lifetime separated from their family against their will (5.66). More
than one quarter of the migrant group experienced an armed conflict (29.34), and 9.2%
migrants and 7.2% non-migrants were confronted with sexual abuse.
Emotional and behavioural problems
Insert table 3
Table 3 shows the scores of both groups on the subscales of the questionnaires on emotional
and behavioural problems. If we use the threshold of 1.75 for the subscales anxiety,
depression and internalising problems of the HSCL-37A, 434 migrants (36%) and 264
(43.9%) non-migrants score within the clinical range on the anxiety subscale. Also on the
subscale depression, a high proportion of both groups score above the threshold: 468 (39.4%)
of the migrants and 246 (40.9%) of the non-migrant group. On the subscale internalising
problems, 457 (38.8%) migrants and 264 (43.9%) non-migrants have a score within the
clinical scores. As the subscale externalising problems of the HSCL-37A is not yet validated,
we cannot apply this threshold.
On the total problem scale of the SDQ, 142 (12.4%) migrants have a score in the borderline
range (16-19) and 84 (7.3%) in the clinical range (20-40). For non-migrants, this is
respectively 101 (16.9%) and 129 (7.4%).
The subscale emotional problems of the SDQ reveals that of the migrant group, 90 (7.8%)
have a score in the borderline range (6) and 130 (11.2%) a clinical score (7-10). 49 (8.2%)
non-migrants have a borderline score on this subscale and 55 (9.2%) score within the clinical
On the subscale conduct problems, 103 (8.9%) migrants score within the borderline range (4),
and 88 (7.6%) in the clinical range (5-10). For the non-migrant group, this is respectively 49
(8.2%) and 58 (9.7%).
The scores on the hyperactivity subscale reveal that 85 (14.2%) adolescents of the non-
migrant group score in the borderline range (6) and 120 (20.1%) in the clinical range (7-10);
for the migrant group, this is respectively 70 (6.1%) and 54 (4.7%).
Concerning the subscale peer problems, 325 (28.2%) migrants have a score within the
borderline range (4-5), and 87 (7.5%) within the clinical range (6-10). 91 (15.2%) non-
migrants have a borderline score and 24 (4%) score in the clinical range.
Of the migrant group, 92 (8%) score within the borderline range (5), and 86 (7.4%) within the
clinical range (0-4) of the subscale pro-social behaviour; 39 non-migrants (6.5%) have
borderline scores, and 39 (6.5%) clinical scores.
On the impact scale, which gives an idea of the impact of the problems on the adolescents’
daily functioning, 653 (66%) migrants and 415 (70%) non-migrants state their problems have
no impact on their daily functioning; 93 (9.4%) and 72 (12.1%) adolescents of both groups
state their problems have some impact, and the problems of 244 (24.6%) migrants and 106
(17.9%) non-migrants have a great impact.
The ZIL-A scores show the levels of post-traumatic stress in both groups. For all subscales,
migrants have higher scores than their non-migrants peers. With a threshold of 52 on the total
ZIL-score, 139 (15.7%) migrants and 40 (6.9%) non-migrants score within the clinical range.
Differences between both groups, and influence of gender, age, living situation and
Using MANCOVA, significant effects are found for group (F(12,1151)=5.23, P<0.001;
²=0.05), gender (F(12,1151)=3.80, P<0.001, ²=0.04), living situation (both parents, one
parents, and other) (F(24,2302)=1.62, P<0.04, ²=0.02), and total number of traumatic events
(F(12,1151)=14.06, P<0.001, ²=0.13). Interaction effects are found for group with gender
(F(12,1151)=2.25, P<0.01, ²= 0.02), group with age (F(60,5393)=1.62, P<0.003, ²=0.017),
group with total number of traumatic events (F(12,1151)=3.34, P<0.001, ²=0.03), gender
with age (F(60, 5393)=1.51, P<0.008, ²=0.02), gender with living situation
(F(24,2302)=1.21, P<0.02, ²=0.01), and gender with total number of traumatic events
(F(12,1151)=2.29, P<0.008, ²=0.02).
Non-migrants have higher scores than migrants on the anxiety (1.77 v. 1.66; F=4.13, d.f.=1,
P<0.05) and externalising subscale (HSCL-37A) (1.56 v. 1.28; F=13.28, d.f.=1, P<0.001),
and on the hyperactivity subscale (SDQ) (4.75 v. 3.04; F=16.78, d.f.=1, P<0.04). On the
contrary, migrants have more peer problems (SDQ) (2.86 v. 2.09; F=18.35, d.f.=1, P<0.001),
and higher avoidance scores (ZIL-A) (16.44 v. 14.81; F=4.52, d.f.=1, P<0.04). No significant
differences between migrants and non-migrants are found on the other subscales.
Girls have more anxiety symptoms (HSCL-37A) (1.78 v. 1.65; F=6.52, d.f.=1, P<0.02), more
emotional problems (SDQ) (3.83 v. 2.75; F=6.72, d.f.=1, P<0.02), and higher avoidance
scores (ZIL-A) (16.10 v. 15.14, F=5.08, d.f.=1, P<0.03); boys have less pro-social behaviour
(SDQ) (7.08 v. 8.14; F=24.53, d.f.=1, P<0.001). Age only influences pro-social behaviour
(SDQ) (F=2.29, d.f.=5, P<0.05), with the lowest scores for 14-years olds.
The living situation of adolescents results in significant differences in anxiety symptoms
(HSCL-37A) (both parents: 1.66, one parent: 1.66, other: 1.83; F=6.52, d.f.=2, P<0.02),
conduct problems (SDQ) (2.10 v. 2.31 v. 2.42; F=3.45, d.f.=2, P<0.04), and pro-social
behaviour (SDQ) (7.83 v. 7.60 v. 7.41; F=6.69, d.f.=2, P<0.002). The total number of
traumatic experiences (SLE) has a significant effect on all subscales (hyperactivity and pro-
social behaviour: P<0.002, peer problems: P<0.004, other subscales: P<0.001).
Interaction effects between group and gender on the externalising subscale (HSCL-37A)
(F=5.35, d.f.=1, P<0.03) and on pro-social behaviour (SDQ) show greater differences
between non-migrant boys and girls than between migrant boys and girls. Between group and
age, interaction effects are found on externalising (HSCL-37A) (F=2.96, d.f.=5, P<0.02) and
peer problems (SDQ) (F=2.73, d.f.=5, P<0.02): for externalising problems, we found for 15-
years olds a decrease in the migrant group and an increase in the non-migrant group; for peer
problems, 14-years old migrants show a decrease, while non-migrants show an increase.
The interaction effect between group and living situation on conduct problems (SDQ)
(F=4.87, d.f.=2, P<0.009) reveals that migrants have the lowest conduct problems scores
when living with one parent, while for non-migrants, the lowest scores are for those living
with both parents.
The interaction effect between gender and age on depression symptoms (HSCL-37A)
(F=3.18, d.f.=5, P<0.008) shows that 16-years old girls have more and boys less depression
symptoms compared to 15-years old, and 18-years old girls show a decrease and boys an
increase compared to 17-years olds. For externalising symptoms (HSCL-37A), 17-years old
girls have higher and boys lower scores compared to 16-years olds (F=2.23, d.f.=5, P<0.05).
On the emotional subscale (SDQ), for boys, the 14-years olds have the lowest scores and the
15-years olds the highest scores, whereas for girls, 15-years olds have the lowest scores and
16-years olds the highest (F=3.15, d.f.=5, P<0.009). On the impact subscale (SDQ), 16-years
old boys have much lower scores compared to 15-years olds, whereas this decrease is much
smaller for girls (F=3.04, d.f.=5, P<0.02).
The interaction effect between gender and living situation on depression symptoms (HSCL-
37A) (F=3.81, d.f.=2, P<0.03) reveals a high increase for boys who don’t live with one or
both parents, compared to a smaller increase for girls in this living situation. For emotional
symptoms (SDQ), girls living with one parent have less emotional problems than those living
with both parents, whereas boys show an increase (F=3.07, d.f.=2, P<0.05).
Influence of trauma, gender, age, and living situation in the migrant group
Insert table 4
Table 4 shows the influence of different variables on the emotional and behavioural problems
in migrant adolescents. The number of traumatic events the migrant adolescents experienced
has a significant influence on the development of all emotional and behavioural problems,
except on the subscales hyper-activity and pro-social behaviour (SDQ).
Gender also influences the emotional problems: girls are more likely than boys to have
anxiety (OR 1.79) and depression (2.12) symptoms (HSCL-37A), and emotional problems
(SDQ) (2.21). On the contrary, girls have less peer problems (0.53) and problems in pro-
social behaviour (0.53).
The age of the adolescent only has small influences on behavioural and emotional problems:
16-years old migrants are more likely to have depression symptoms (HSCL-37A) (1.69), and
emotional problems (SDQ) (1.94), compared to migrants up to 13 years of age; also 15- (2.21)
and 17-years olds (1.97) may have more emotional problems (SDQ). 16- (0.52) and 17-years
olds (0.51) are less likely to have problems in pro-social behaviour (SDQ).
Finally, migrant adolescents who live alone in Belgium are a high risk group for the
development of anxiety (1.82) and depression (2.84) (HSCL-37A), and emotional problems
(SDQ) (1.95), compared to migrants living with both parents. For post-traumatic stress, they
have a relative risk of 2.94, compared to those living with both parents. On the contrary, they
have less conduct problems (SDQ) (0.29).
Relationships between type of trauma and post-traumatic stress
Insert table 5
The type of trauma an adolescent experienced may influence the level of post-traumatic
stress. In table 5, differences between the migrant and non-migrant adolescents concerning the
influence of the trauma type on a full symptom PTSD diagnose is shown. For migrants, there
is little difference between the trauma types in their effect on the post-traumatic stress level,
and only two types have a significant influence: changes in family life (OR 1.73) and physical
abuse (2.03). For non-migrant adolescents, differences are much greater, with physical (4.69)
and sexual abuse (5.27) as important traumatic types in the development of post-traumatic
stress, even as separation from the family (3.84).
People fleeing from war or persecution, from poverty or disasters is a worldwide phenomenon
of all time. Also minors leave their home countries, and migrate to build a new life in an
unknown country. This experience of migration can place young people at risk for the
development of emotional and behavioural problems. Four factors might influence this: At
first, all migrating children and adolescents share the experiences of loss and separations: the
loss of one’s home and belongings, familiar environment, friends and relatives, usual patterns
of family life and education, social network and cultural context (Ajdukovic & Ajdukovic,
1993). The escape breaks up the existing family system, and refugees loose not only their
social heritage, but also their social identity, bound to their origin. And, all this happens
during adolescence, the critical transition period for identity development, which involves that
migration may complicate the basis adolescent processes, such as separation from parents,
choice of social role, and the search for an adult identity (Ajdukovic, 1998). Thirdly, refugee
and migrant adolescents may have to deal with direct or indirect traumatic experiences they
have encountered before or during the escape from their homeland (Berman, 2001). Finally,
they have to start a new life in an unknown country, often without knowing if they can stay or
not, they have to adapt to a new society, build a new social network, learn a new language and
new cultural practices, they may encounter teasing and bullying, and many migrant families
are forced into marginalized positions (Sack, 1998).
Emotional and behavioural problems
In this study, we found high rates of both migrant and Belgian adolescents having clinical
scores of anxiety and depression (HSCL-37A): 36% of the migrant group and almost 44% of
the non-migrants have serious anxiety problems, and around 40% of both groups have severe
depression symptoms. On the emotional problems subscale of the SDQ, lower percentages of
emotional problems were found, with 19% of the migrant group and 17.4% of the non-
migrants having borderline or clinical scores. A possible explanation for this difference may
be that the adult-oriented thresholds of the HSCL-37A should be adapted for the use in an
Belgian adolescents have more anxiety symptoms than non-migrants; for depression and
emotional symptoms, no differences between the two groups were found. This latter finding is
in accordance with some research (Munroe-Blum, Boyle, Offord, et al, 1989; Klimidis,
Stuart, Minas, et al, 1994; Rousseau & Drapeau, 1998), while the first finding corresponds
with studies reporting migrants having fewer emotional problems than non-migrants (Davies
& McKelvey, 1998; Sam, 1994). On the contrary, some studies reported higher emotional
problems in migrant children and adolescents compared to native-born peers (Bengi-Arslan,
Verhulst, van der Ende, et al, 1997; Darwish Murad, Joung, van Lenthe, et al, 2003; Diler,
Avci & Seydaoglu, 2003). We should take into account that the problems in the migrant group
may be an underestimation of the real figures, because migrants may feel less comfortable
reporting behaviours that might be perceived as deviant out of their cultural backgrounds, and
individuals and families are taught to keep emotional and behavioural problems to themselves
(Davies & McKelvey, 1998).
Concerning behavioural problems, 16.5% of the migrants and 17.9% of the non-migrant
adolescents have important conduct problems, and non-migrants have significant more
externalising problems. This is in accordance with different studies (e.g., Darwish Murad,
Joung, van Lenthe, et al, 2003; Stevens, Pels, Bengi-Arslan, et al, 2003). Belgian adolescents
also have higher hyperactivity scores than migrants, with 34.3% Belgians and 10.8% migrants
having borderline or clinical scores. A possible explanation for these findings may be the
striving of migrants to have a better future by attempting to obtain good academic
Further on, migrants have significant more peer problems than non-migrant adolescents
(respectively 35.7% and 19.2% have borderline or clinical scores), which is in accordance
with the research of Darwish Murad, Joung, van Lenthe, et al (2003), and migrants have more
borderline or clinical scores on the pro-social behaviour subscale compared to non-migrants
(respectively 15.4% and 13%). These findings may be due to the possible acculturation
difficulties newly arrived migrants may encounter.
9.4% migrant and 12.1% non-migrant adolescents state their problems have some impact on
their daily functioning, and one quarter of the migrant group and almost one fifth of the non-
migrants say their problems have a great impact on their daily functioning.
Traumatic experiences and post-traumatic stress
Migrant adolescents experienced significant more traumatic events than Belgians (a mean of
3.6 compared to 2.7). The death of a loved one was the most experienced traumatic event for
both groups. More than one quarter of the migrants experienced a war or armed conflict and
almost one quarter was once separated from their family. Most of the traumatic events, except
four, were experienced more by migrants; the life event ‘experienced a stressful life event
where he/she thought that someone else was in danger’ was experienced more by non-
In this study, migrant had significant higher avoidance scores than non-migrant adolescents,
and more migrants (15.7%) than non-migrants (6.9%) had clinical PTSD scores. These
percentages are comparable to some studies (Seedat, Nyamai, Njenga, et al, 2004; Servan-
Schreiber, Le Lin & Birmaher, 1998), but lower than in other studies, where between 20 and
98% of the migrant and refugee children and adolescents met clinical PTSD levels (e.g.,
Allwood, Bell-Dolan & Husain, 2002; Goldstein, Wampler & Wise, 1997). As already stated,
this may be due to the fact that migrants may under report emotional problems.
The type of traumatic events that lead to a full symptom PTSD diagnosis was different for
migrants and non-migrants: for migrants changes in family life (OR 1.73) and physical abuse
(2.03) lead to higher risks on PTSD, while for non-migrants, this was separation from family
(3.84), physical (4.69) and sexual abuse (5.27). This important influence of sexual abuse is
also supported in other studies (Seedat, Nyamai, Njenga, et al, 2004).
Influence of number of traumatic experiences, gender, age, and living situation for both
For both groups, the number of traumatic experiences is an important influencing factor in the
development of emotional and behavioural problems. This is in accordance with many other
studies (e.g., Seedat, Nyamai, Njenga, et al, 2004; Weine, Becker, McGlashan, et al, 1995).
In both groups, girls have more anxiety symptoms, emotional problems and avoidance
symptoms compared to boys. On the contrary, boys have less pro-social behaviour, although
this difference between girls and boys is greater in the non-migrant group. Moreover, in the
migrant group, girls have more depression symptoms and less peer problems than boys. All
these findings are supported in other studies (e.g., Allwood, Bell-Dolan & Husain, 2002;
Diler, Avci & Seydaoglu, 2003; Verhulst, Achenbach, van der Ende, et al, 2003). The age of
the adolescent only has small influences on the development of emotional and behavioural
problems. Also in literature, there are little consistent results about the influence of age on the
development of behavioural and emotional symptoms (Berman, 2001).
The living situation of adolescents influences their emotional and behavioural problems.
Migrant and non-migrant adolescents who do not live with one or both parents are at risk for
the development of anxiety symptoms, conduct problems and less pro-social behaviour.
Moreover, unaccompanied migrant adolescents are a high risk group for the development of
anxiety, depression, emotional problems and post-traumatic stress. On the contrary, they have
less conduct problems than accompanied migrant adolescents. This strongly supports the
protective influence of the availability of a parent, which is also shown in many other studies
(e.g., Derluyn, Broekaert, Schuyten, et al, 2004; Sourander, 1998).
Limitations of the research
A first limitation of this study is that, although a great part of the population of the 11- to 18-
years old, non-Dutch speaking newcomers in Flanders was included, the study group remains
a diverse group, composed of adolescents of 93 different countries, and differing from the
cultural background of the Belgian control group.
Besides the impact subscale of the SDQ, we did not assess the level of functional impairment
produced by the emotional and behavioural symptoms, which might result in a over-diagnosis
of disorders (Guarnaccia & Lopez, 1998). The questionnaires do not provide a qualitative
understanding of how adolescents are affected by their experiences, and the symptoms asked
for are only one dimension of the possible reactions to migration and traumatic events
(Berman, 2001). We also did not use clinical interviews, which may be helpful to obtain a
greater richness of information from participants than self-report measures alone (Barrett,
Moore & Sonderegger, 2000), and we only questioned the adolescent’s perception on his
problems. We do not have information about the legal status of the adolescents, about family
characteristics, and about their current economic and social living situation. We also do not
know how the reactions of the adolescents to the stressors may be mediated by the nature,
exposure and duration of traumatic stressors, by the meaning or interpretation attached to
these events by their community and family, and by individual characteristics of the
adolescent, such as intelligence, pre-migration status, and coping strategies.
Certain adolescents may have experienced difficulties to understand the questionnaires,
because they were not available in all languages, or some adolescents were illiterate. The
questionnaires are not culturally validated for the various ethnic groups in which they were
used, and the likelihood of cultural response bias to questionnaire items cannot be excluded.
As already stated, migrant adolescents might feel less comfortable reporting behaviours that
might be perceived as deviant, since mental illness is stigmatised in some cultures, and
individuals and families are thus taught to keep emotional and behavioural problems to
themselves (Davies & McKelvey, 1998). Moreover, it is important to take into account the
difficulties when using psychological concepts and assessment instruments in other cultures,
because ethno-cultural variation may decrease their validity, and the fact that symptoms and
signs can be identified with standardized questionnaires no guarantee that they mean the same
thing in those settings (Bracken, Giller & Summerfield, 1995).
These study shows the possibilities of using instruments as screening tool for the assessment
of emotional and behavioural problems in migrant adolescents, although these instruments do
not lead to a full clinical diagnosis of mental disorder.
Since many studies have shown that emotional and behavioural problems reported by
adolescents may predict the onset of DSM-IV mood and anxiety disorders in adulthood
(Roza, Hofstra, van der Ende, et al, 2003), this study underscores the importance of early
detection of these problems and, if needed, early interventions. However, although we found
high rates of emotional and behavioural problems and traumatic experiences in migrant
adolescents, this does not involve that these youngsters do not continue to function well in
their daily life (Mollica, Poole, Son, et al, 1997). Therefore, we are convinced that it is very
important to avoid stigmatisation of these adolescents, and consider them as normal people
who have to live in abnormal circumstances (Papadopoulos, 1999).
We thank all participating adolescents and schools.
This study was supported by a PhD Bursary of the Special Research Fund of Ghent
Unaccompanied minors are a high risk group for the development of emotional
problems and post-traumatic stress.
Attention should be given to early detection and, if needed, intervention to prevent the
development of severe emotional and behavioural problems in migrant adolescents.
Stigmatisation of migrant adolescents must be avoided, and they should be considered
as normal minors who live in abnormal circumstances
Limitations of the study
The self-report questionnaires used in this study were not culturally validated and
were not translated in all languages.
We have limited information about the background and current living situation of the
adolescents, and we only included the view of the adolescent himself.
The symptoms asked for in this study are only one dimension of the possible reactions
to migration and traumatic events of adolescents.
Ajdukovic, M. (1998) Displaced adolescents in Croatia: Sources of stress and posttraumatic
stress reaction. Adolescence, 33, 209-217.
Ajdukovic, M. & Ajdukovic, D. (1993) Psychological well-being of refugee children. Child
Abuse and Neglect, 17, 843-854.
Allwood, M. A., Bell-Dolan, D. & Husain, S. A. (2002) Children's trauma and adjustment
reactions to violent and nonviolent war experiences. Journal of the American Academy
of Child and Adolescent Psychiatry, 41, 450-457.
Barrett, P. M., Moore, A. F. & Sonderegger, R. (2000) The FRIENDS Program for Young
Former-Yugoslavian Refugees in Australia: A Pilot Study. Behaviour Change, 17,
Bean, T. (2000) Measuring the effects of traumatic stress on immigrant and refugee
adolescents. Doctoraalscriptie. Leiden: Leiden University - Department of Clinical
Child and Youth Psychology.
Bean, T., Spinhoven, P., Eurelings-Bontekoe, E. H. M., et al (in press) Refugee Adolescent
Screening Instrument, user's manual.
Bengi-Arslan, L., Verhulst, F. C., van der Ende, J., et al (1997) Understanding childhood
(problem) behaviors from a cultural perspective: comparison of problem behaviors
and competencies in Turkish immigrant, Turkish and Dutch children. Social
Psychiatry and Psychiatric Epidemiology, 32, 477-484.
Berman, H. (2001) Children and war: current understandings and future directions. Public
Health Nursing, 18, 243-252.
Bracken, P. J., Giller, J. E. & Summerfield, D. (1995) Psychological responses to war and
atrocity: the limitations of current concepts. Social Science & Medicine, 40, 1073-
Browne, M. W. & Cudeck, R. (1993) Alternative ways of assessing model fit. In Testing
structural equation models (eds K. A. Bollen & J. S. Long), pp. 136-162. Newbury
Darwish Murad, D., Joung, I. M. A., van Lenthe, F. J., et al (2003) Predictors of self-
reported problem behaviours in Turkish immigrant and Dutch adolescents in the
Netherlands. Journal of Child Psychology and Psychiatry, 44, 412-423.
Davies, L. C. & McKelvey, R. S. (1998) Emotional and behavioural problems and
competencies among immigrant and non-immigrant adolescents. Australian and New
Zealand Journal of Psychiatry, 32, 658-665.
Derluyn, I., Broekaert, E., Schuyten, G., et al (2004) Post-traumatic stress in former
Ugandan child soldiers. Lancet, 363, 861-863.
Derogatis, L. R., Lipman, R. S., Rickels, K., et al (1974) The Hopkins Symptom Checklist
(HSCL): A self-report symptom inventory. Behavioral Science, 19, 1-15.
Diler, R. S., Avci, A. & Seydaoglu, G. (2003) Emotional and behavioural problems in
migrant children. Swiss Medical Weekly, 133, 16-21.
Goldstein, R. D., Wampler, N. S. & Wise, P. H. (1997) War experiences and distress
symptoms of Bosnian children. Pediatrics, 100, 873-878.
Goodman, R. (1997) The Strengths and Difficulties Questionnaire: A research note. Journal
of Child Psychology and Psychiatry and Allied Disciplines, 38, 581-586.
---- (1999) The extended version of the strengths and difficulties questionnaire as a guide to
child psychiatric caseness and consequent burden. Journal of Child Psychology and
Psychiatry and Allied Disciplines, 40, 791-799.
---- (2001) Psychometric properties of the strengths and difficulties questionnaire. Journal of
the American Academy of Child and Adolescent Psychiatry, 40, 1337-1345.
Goodman, R., Ford, T., Simmons, H., et al (2000) Using the Strengths and Difficulties
Questionnaire (SDQ) to screen for child psychiatric disorders in a community sample.
British Journal of Psychiatry, 177, 534-539.
Goodman, R., Meltzer, H. & Bailey, V. (1998) The Strengths and Difficulties
Questionnaire: A pilot study on the validity of the self-report version. European Child
& Adolescent Psychiatry, 7, 125-130.
Guarnaccia, P. J. & Lopez, S. (1998) The mental health and adjustment of immigrant and
refugee children. Child and Adolescent Psychiatric Clinics of North America, 7, 537-
Hovens, J., Bramsen, I. & van der Ploeg, H. (2000) Handleiding Zelfinventarisatielijst.
Lisse: Swets Test Publishers.
Kleijn, W. C., Hovens, J. E. & Rodenburg, J. J. (2001) Posttraumatic stress symptoms in
refugees: Assessments with the Harvard Trauma Questionnaire and the Hopkins
Symptom Checklist-25 in different languages. Psychological Reports, 88, 527-532.
Klimidis, S., Stuart, G., Minas, I. H., et al (1994) Immigrant status and gender effects on
psychopathology and self-concept in adolescents: a test of the migration-morbidity
hypothesis. Comprehensive Psychiatry, 35, 393-404.
Lavik, N. J., Hauff, E., Skrondal, A., et al (1996) Mental disorder among refugees and the
impact of persecution and exile: Some findings from an out-patient population. British
Journal of Psychiatry, 169, 726-732.
Lavik, N. J., Laake, P., Hauff, E., et al (1999) The use of self-reports in psychiatric studies
of traumatized refugees: validation and analysis of HSCL-25. Nordic Journal of
Psychiatry, 53, 17-20.
Mollica, R., Poole, C., Son, L., et al (1997) Effects of war trauma on Cambodian refugee
adolescents' functional health and mental health status. Journal of the American
Academy of Child Psychiatry, 36, 1098-1106.
Mollica, R. F., Wyshak, G., de Marneffe, D., et al (1986) Hopkins Symptom Checklist-25
manual: Cambodian, Laotian, and Vietnamese versions. Washington, D.C.: U.S.
Office of Refugee Resettlement.
Mullick, M. S. I. & Goodman, R. (2001) Questionnaire screening for mental health
problems in Bangladeshi children: a preliminary study. Social Psychiatry and
Psychiatric Epidemiology, 36, 94-99.
Munroe-Blum, H., Boyle, M. H., Offord, D., et al (1989) Immigrant children: psychiatric
disorder, school performance and service utilization. American Journal of
Orthopsychiatry, 59, 510-519.
Papadopoulos, R. (1999) Working with Bosnian medical evacuees and their families:
Therapeutic dilemmas. Clinical Child Psychology and Psychiatry, 4, 107-120.
Parloff, M. B., Kelman, H. C. & Frank, J. D. (1954) Comfort effectiveness, and self-
awareness as criteria for improvement in psychotherapy. American Journal of
Psychiatry, 3, 343-351.
Perez Foster, R.-M. (2001) When immigration is trauma: guidelines for the individual and
family clinician. American Journal of Orthopsychiatry, 71, 153-170.
Rousseau, C. & Drapeau, A. (1998) Parent-child agreement on refugee children's psychiatric
symptoms: a transcultural perspective. Journal of the American Academy of Child and
Adolescent Psychiatry, 37, 629-636.
Roza, S. J., Hofstra, M. B., van der Ende, J., et al (2003) Stable prediction of mood and
anxiety disorders based on behavioral and emotional problems in childhood: A 14-
year follow-up during childhood, adolescence, and young adulthood. American
Journal of Psychiatry, 160, 2216-2121.
Sack, W. H. (1998) Multiple forms of stress in refugee and immigrant children. Child and
Adolescent Psychiatric Clinics of North America, 7, 153-167.
Sam, D. L. (1994) The psychological adjustment of young immigrants in Norway.
Scandinavian Journal of Psychology, 35, 240-253.
Sandanger, I., Moum, T., Ingebrigtsen, G., et al (1999) The meaning and significance of
caseness: the Hopkins Symptom Checklist-25 and the Composite International
Diagnostic Interview II. Social Psychiatry and Psychiatric Epidemiology, 34, 53-59.
Seedat, S., Nyamai, C., Njenga, F., et al (2004) Trauma exposure and post-traumatic stress
symptoms in urban African school. Survey in Cape Town and Nairobi. British Journal
of Psychiatry, 184, 169-175.
Servan-Schreiber, D., Le Lin, B. & Birmaher, B. (1998) Prevalence of posttraumatic stress
disorder and major depressive disorder in Tibetan refugee children. Journal of the
American Academy of Child and Adolescent Psychiatry, 37, 874-879.
Sourander, A. (1998) Behavior problems and traumatic events of unaccompanied refugee
minors. Child Abuse and Neglect, 22, 719-727.
Stevens, G., Pels, T., Bengi-Arslan, L., et al (2003) Parent, teacher and self-reported
problem behavior in The Netherlands - Comparing Moroccan immigrant with Dutch
and with Turkish immigrant children and adolescents. Social Psychiatry and
Psychiatric Epidemiology, 38, 576-585.
Verhulst, F. C., Achenbach, T. M., van der Ende, J., et al (2003) Comparison of problems
reported by youths from seven countries. American Journal of Psychiatry, 160, 1479-
Weine, S. M., Becker, D. F., McGlashan, T. H., et al (1995) The psychiatric consequences
of "ethnic cleansing": clinical assessments and trauma testimonies of newly resettled
Bosnian refugees. American Journal of Psychiatry, 152, 536-542.
Name and addresses of the authors:
Ilse Derluyn, MEduc
Department of Orthopedagogics
H. Dunantlaan 2
Eric Broekaert, PhD
Department of Orthopedagogics
H. Dunantlaan 2
Gilberte Schuyten, PhD
Department of Data-Analysis
H. Dunantlaan 1
Department of Orthopedagogics
H. Dunantlaan 2
Phone: (+32) (0) 9 264 63 63
Fax: (+32) (0) 9 264 64 91
E-mail : Ilse.Derluyn@UGent.be