Abstract
Currently, there are approximately 10.8 million child refugees worldwide. Youth living in refugee camps face a wide range of difficulties placing them at risk for trauma exposure and negative mental health outcomes. However, little is known about the mental health functioning of these youth. The present review provides a systematic review of mental health outcomes for refugee/displaced youth residing in refugee camps. Twenty studies were included in the present review. Among these studies, the prevalence of mental health disorders varied greatly with some studies reporting null effects for disorders like posttraumatic stress disorder and others reporting prevalence as high as 87%. Levels of anxiety, somatic symptoms, depression, and aggression also varied across studies. The results point to the significant need for more research on the mental health of youth residing in refugee camps. Despite the wide range of measurement approaches, the evidence points to a fairly consistent finding of a range of maladjustment problems for youth living in refugee camps. Implications for improving the methodology for investigating mental health are discussed.
Key Points of the Research Review
Youth living in refugee camps are exposed to unique stressors such as living in active war zones; higher levels of physical and sexual violence; and minimal access to shelter, food, water, and education.
Across the studies reviewed, a multitude of different assessments were utilized to measure the prevalence of mental health problems for youth in refugee camps, ranging from author-created measures to standardized assessments.
Translation methods of assessments varied across the studies ranging from lack of reporting on translation procedures to a multimethod translation process that ensured accuracy.
Prevalence rates of mental health disorders varied across the refugee camps reviewed.
The United Nations High Commissioner for Refugees (UNHCR, 2016) Global Trends Report estimates that the number of refugees and internally displaced people exceeds 65 million; 21.3 million of these individuals are refugees. Although 414,600 refugees returned home in 2013, roughly 32,000 people are forced to flee their country of origin everyday, which includes approximately 13,000 youth (UNHCR, 2013). Roughly 6.7 million refugees have been displaced for more than 5 years making camp life more of a permanent living arrangement than a temporary experience (UNHCR, 2016). Furthermore, in 2015, at least 51% (i.e., 10.8 million) of refugees were children, which represents the highest number of displaced youth in over a decade (UNHCR, 2016). Although the number of youth living in refugee camps continues to increase, little is known about the mental health of these youth, many of whom spend the majority of their childhoods in refugee camps (Feldman, 2007). The present review seeks to systematically explore and describe the literature base regarding the mental health outcomes of these youth.
A refugee is defined as anyone who has a well founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear is unwilling to avail himself of the protection of that country. (UNHCR, 2015) forcibly displaced within their countries of origin or habitual residence but who have not crossed an internationally recognized State boarder. Individuals may be internally displaced as a result of armed conflicts, situations of generalized violence, violations of human rights or natural or human—made disasters. (UNHCR, 2015)
Youth, in particular, face a wide range of difficulties living in refugee camps including inadequate standards of water availability, malnutrition, inadequate shelters, and lack of education (United Nations Development Program; Bruijn, 2009). In addition to growing up with limited resources, the camp environment may be a source of trauma for youth. In an examination of Kenyan refugee camps, Crisp (2000) described domestic violence, sexual abuse, genital mutilation, armed robbery, and violence between national refugee groups as common occurrences within camps. Traumatic experiences may increase the risk for developmental delays, difficulties interacting with peers, and emotion regulation difficulties for youth (New South Wales Refugee Health Center, 2009). Moreover, caregivers, who may also be traumatized by war and displacement, may also have a diminished ability to provide physical and emotional support for their children.
Mental Health of Displaced Youth
Due to the many difficulties of camp life (i.e., minimal access to shelter, food, and water), trauma exposure, the long-term stay, and the large populations of displaced youth, an examination of mental health outcomes for youth in refugee camps is important. Currently, mental health research on youth in refugee camps has examined anxiety, depression, aggression, coping, posttraumatic stress disorder (PTSD), somatic symptoms, and locally defined syndromes (i.e., Betancourt, Speelman, Onyango, & Bolton, 2009; Thabet, Abed, & Vostanis, 2004). Results have varied with regard to the type of mental health problem being investigated as well as the direction of the effects. For example, Muecke and Sassi (1992) found low levels of anxiety among youth in a Thai refugee camp, whereas Yurtbay, Alyanak, Abali, Kaynak, and Durukan (2003) found high levels of anxiety in a Kenyan refugee population. The broad variety of mental health outcomes examined and the variations among findings indicate a need for a systematic review of the knowledge base on mental health for refugee youth. Thus far, previous reviews have focused on displaced status and resettlement broadly rather than youths’ experiences within the camp.
For example, Lustig and colleagues (2004) examined studies conducted between 1990 and 2003 of special refugee populations, such as unaccompanied minors, asylum seekers, and former child soldiers, during their preflight, flight, and resettlement stages of the refugee experience. Findings showed that youth and adolescent refugees who had resettled into a safer environment continued to report moderate to high levels of negative mental health outcomes such as anxiety, depression, paranoia, insomnia, PTSD, and heightened awareness of death. Further, Reed, Fazel, Jones, Panter-Brick, and Stein (2012) reviewed the literature on risk and protective factors of youth who were permanently resettled in low- to middle-income countries and examined the mental health outcomes of youth in the resettlement phases. Similar to findings in Lustig et al. (2004), Reed and colleagues found that exposure to violence and combat was a significant risk factor for negative mental health outcomes.
Finally, Fazel, Reed, Panter-Brick, and Stein (2012) conducted a follow-up review of the Reed et al.’s (2012) study, examining the risk and protective factors of resettled youth, but focused on those who resettled in high-income countries. For this review, Fazel and colleagues only included studies that examined mental health outcomes (e.g., depression, anxiety, PTSD) of youth who were forcibly displaced from their homes and resettled into high-income countries. Consistent with the previous reviews, exposure to violence was predictive of negative mental health outcomes for youth refugees who settled in high-income countries.
Previous literature has addressed mental health in youth during the different phases of displacement for youth; however, no review has focused exclusively on mental health outcomes of youth while they are living in a refugee camp.
The Present Review
The present review sought to address this gap by providing a systematic review of the research on mental health outcomes for youth who were living in refugee camps during the time the study was conducted. Exclusively examining outcomes for youth currently living in refugee camps is necessary due to the unique difficulties experienced in camp life that could result in a wide range of mental health concerns for youth, including depression, anxiety, PTSD, somatoform disorders, and aggression (Brujin, 2009; Toole, 1994). Examining the outcomes of youth who live in refugee camps is becoming increasingly important as the number of youth living in refugee camps continues to increase and the amount of time youth live in the camps continues to lengthen (Feldman, 2007). Furthermore, some displaced families are preferentially accepted into countries for resettlement (i.e., educated families) while others must resort to living in refugee camps; thus, focusing on youth in camps allows for an examination of the most vulnerable populations with the least access to resources (Fazel et al., 2012). Lastly, focusing on studies that assess youth mental health symptoms while at the camp could minimize the problems with retrospective recall data, therefore possibly providing a clearer view of the more immediate mental health outcomes for the multitude of youth living in refugee camps.
Method
Search Strategy and Criteria of Inclusion
Systematic review of four major scientific databases was completed for the current study: PsycInfo, PubMed, Google Scholar, and Sociological Abstracts including from inception of each database to February 2015, using combinations of the following search terms: refugee, displaced, youth, child, refugee camp, mental health, mental health outcomes, anxiety, depression, aggression, PTSD, and negative outcomes. Furthermore, the World Health Organization, United Nations, and Child Rights International Network (CRIN) databases were searched for additional references. If articles appeared to be relevant (i.e., title or key word tags included any of the above search term words), their abstracts were read. Articles that appeared to meet eligibility criteria based on the abstract were read and assessed to determine whether they met criteria to be included in the study. Furthermore, all eligible articles reference lists and authors were checked for other relevant articles, and those studies were subjected to the same selection process. The review includes no restrictions on the year of publication to ensure all possible relevant articles were incorporated.
To be included in the present review, studies had to meet the following criteria: (1) The sample had to be specified as refugee or internally displaced and had to reside in a refugee camp at the time the study was being conducted, (2) had to include assessments of youth age 18 or younger, and (3) explicitly state a focus on the mental health of youth or adolescents (i.e., internalizing or externalizing problems), and the results had to directly assess the prevalence of one or multiple mental health outcomes for youth living in camps.
Search Results
Review of electronic databases produced 418 articles. After initial matching of title or key words to search terms, 85 articles were selected for abstract review. Based on the abstracts, 58 articles were read fully for inclusion. Thirty articles were excluded because they did not assess mental health outcomes (e.g., focus was on coping, risk factors, or family roles) and eight articles were excluded from the review because the analyses did not assess youth currently living in camps;,thus 20 articles met inclusion criteria for the present review.
Second-Order Theme Identification
The 20 articles that met inclusion criteria were each thoroughly reviewed for the mental health outcomes they reported. Each article was also reviewed for information on continent location of the refugee camp, duration in camp, trauma experiences before and during camp stay, age and gender of participants, and participants mother’s mental health outcomes. Mental health outcomes were then compared with each of these factors to see if all the articles taken together as a whole provided any second-order themes on mental health outcomes. These second-order themes are discussed in greater detail in Results section.
Results
Demographic Information and Recruitment Methods
Across the studies, the age of participants ranged from 3 to 22 years old (see Table 1). Most of the studies (15) had an equal gender distribution, while 2 studies had primarily female participants, 1 study had mostly male participants, and 2 studies did not report the gender distribution. Finally, studies assessed refugee camps that were located in Asia (seven studies), Europe (six studies), Africa (four studies), and North America (three studies).
Summary of Demographic Information on Refugee Youth Mental Health Outcomes.
aSummary includes the most common mental health issues reported across studies (i.e., anxiety, PTSD), thus specific and infrequently reported issues such as weeping, eating disorders, nightmares, and so on were not included.
Measures of Mental Health
Reporters
Eight studies in this review used youth self-report to obtain information on mental health symptoms (i.e., Stein, Comer, Gardner, & Kelleher, 1999), while eight studies utilized both parent/caregiver and self-report (i.e., Ajdukovic & Ajdukovic, 1998). Many of the studies that utilized dual reports also used multiple assessments to capture different constructs. For example, Goldstein, Wampler, and Wise (1997) used parent reports to assess the occurrence of war related experiences, whereas the child’s self-report was used to assess the presence of distress symptoms and posttraumatic stress symptoms. Only two studies focused solely on parents’ report of children’s mental health (Ajdukovic & Ajdukovic, 1993; Smith, Sabin, Berlin, & Nackerud, 2009). Two studies used observation (Ekblad, 1993; Irshad & Bano, 2004) and another simultaneously utilized observation, self- and teacher report (Rothe, 2005). Three studies also included physical health outcome measures (Izutsu et al., 2005; Mollica, Poole, Son, Murray, & Tor, 1997; Van Ommeren et al., 2001).
Type of measurement
Within the broad category of mental health symptoms, a multitude of different tools were used, including standardized assessments, author-created study-specific measures, and qualitative interviews (see Table 1 for a list of measures and outcomes).
Translation Method
Most studies were conducted with nonnative English-speaking individuals, so many of the measures had to be translated and the procedures for translation varied across studies. Methods for translation can be categorized into three different types: direct translation, back translation, and translation with pretest/discussion using key informants. Direct translation involves at least one person translating surveys into the native language of the participants. Back translation involves translating the measure into the native language of participants and then having a different fluent speaker translate the measure back to English (or the original language). Due to multiple reviews of the text for accuracy in two languages (new language and original language), this method is likely more rigorous than the direct translation method. Finally, the key informant method ensures validity by verifying that the translation captures the intended meaning for the participants. The key informant method typically involves a direct translation of a questionnaire (or a back translation), which is then given to key informants who ensure conceptualization of locally relevant constructs.
Eight of the studies in this review did not describe the translation process used for their study, whereas six studies mentioned a direct or one-way translation method. Ahmad, Sofi, Sundelin-Wahlsten, and von Knorring (2000) used a direct translation method in which the English surveys were translated in real time as they were being administered to participants. Five studies provided a description of their back-translation procedures. For example, the principal investigator and three Sudanese translators fluent in Arabic and English conducted translations in the Morgos, Worden, and Gupta’s (2008) study. Questionnaires were translated from English to Arabic and items were reviewed for contextual and cultural appropriateness, and then two independent translators translated the questionnaires back to English to ensure accuracy.
Peltzer (1999) was the only study in this review to use the key informant validation process in addition to back translation. After back translation was complete, both of the translation groups had to reach consensus on specific concepts and then the surveys were pretested in a local refugee population and advice was sought from the local authorities in the community for understandability. Surveys in the Peltzer’s (1999) study were administered twice in the pilot study and had a reliability coefficient above .9 on all questionnaires.
Mental Health Outcomes
Although all studies included in this review focused on mental health outcomes, the types of maladjustment examined varied. Most of the studies in some way assessed for trauma symptoms or PTSD (11), anxiety and depression (13), aggression (6), somatic symptoms (7), and locally defined syndromes (2). A majority of studies focused on examining multiple mental health outcomes or examined mental health outcomes in general (11), whereas three studies stated a specific focus on anxiety, one study only evaluated aggression, three studies specifically examined PTSD, and two studies had a specific focus on locally defined mental health disorders.
PTSD symptoms
Prevalence rates for PTSD symptoms were mixed. Miller (1996) found no significant levels of PTSD symptoms in a study of Guatemalan Mayan youth residing in Mexican refugee camps, whereas Paardekooper, de Jong, and Hermanns (1999) found Sudanese refugee children had higher levels of PTSD symptoms than Ugandan nonrefugee comparison group. Additionally, Ahmad, Sofi, Sundelin-Wahlsten, and von Knorring (2000) measured PTSD in 45 Iraqi Kurdistan families randomly selected from two refugee camps. Results suggested that 87% of children and 60% of caregivers met criteria for PTSD. Childhood trauma scores and duration of captivity predicted youth PTSD, while maternal PTSD did not predict youth’s PTSD diagnosis. Other studies found that the duration of time spent living in camps was associated with reductions of PTSD symptom severity (Ajdukovic & Ajdukovic, 1998; Stein et al., 1999). One such study found that this reduction was stronger for boys than it was for girls (Stein et al., 1999).
In another study examining comorbidity of PTSD and depression among 403 Palestinian children, 21.1% of children reported symptoms that could be defined as a mild PTSD reaction, 52.6% with moderate, 22.9% with severe, and 1% with very severe reactions (Thabet et al., 2004). Results indicated that 95 children (23.9%) scored in the clinical range for PTSD. Thabet and colleagues also found a positive relationship between depressive and PTSD symptoms, in which children who scored above the clinical cutoff for PTSD also had significantly higher depression scores than those children with scores below the clinical PTSD range. Exposure to traumatic events was a strong predictor of depression when controlling for PTSD scores, whereas the association between traumatic events and youth PTSD scores was weak when controlling for depression scores. The authors concluded the association between depression and PTSD may be partly due to symptom overlap.
Rothe (2005) examined three cohorts of Cuban refugee children staying in camps at Guantanamo Bay. One cohort of 285 youth was considered a clinical population because they had sought psychiatric services in either of the two hospitals adjacent to the refugee camp. A second cohort of 79 Cuban youth refugees left Cuba through a route that is longer and more traumatic than the typical route taken, in which they were confined to detention camps, exposed to abuse, and then transported to the Guantanamo U.S. Naval base. A third cohort of 87 Cuban youth refugees had been assessed 6 months after leaving the Guantanamo refugee camp. In the refugee clinical population, the majority of children (57%) met criteria for PTSD. In the second cohort of children, 75% reported fear of recalling traumatic events, 81% reported a fear the events would repeat, 52% reported headaches, 61% reported weight loss, 66% reported generalized anxiety, and 56% reported insomnia. Compared to the refugee clinical population and the second cohort, the third cohort of children (who had been recently released from a camp) had lower levels of PTSD, in which only 25% met criteria for PTSD.
Anxiety and depressive symptoms
Similar to the results on PTSD symptoms, results were mixed for the studies that assessed for anxiety; with some studies finding worse anxiety symptomology from refugee youth, whereas others reported that refugee youth were faring better than resettled youth. Muecke and Sassi (1992) assessed for anxiety and emotional distress symptoms for their sample of Cambodian refugees in Thai camps and Thai youth resettled in the United States. Results indicated that adolescents in the refugee camp reported lower levels of anxiety (M = 1.23) than Thai children who had resettled in the United States (M = 1.75). Furthermore, both groups scored very low on the Life Events Checklist, an assessment of stressful and anxiety provoking life events, with the resettled Thai youth reporting (M = 6.2) slightly higher scores than the refugee youth (M = 5.1). These results differ from those of Yurtbay and colleagues (2003) who compared 250 Albanian youth residing in refugee shelters and 118 nonrefugee youth. Results showed that 35% of the refugee children experienced mild levels of anxiety, 15% of youth experienced moderate levels, and 4.7% of youth experienced high levels of trait anxiety. Comparatively, only 9% of youth in the control group experienced mild levels of anxiety, 6% of youth experienced moderate levels, and none of the nonrefugee youth experienced high levels of trait anxiety. Furthermore, the refugee group endorsed significantly more depression and negative self-feelings than the control sample.
Results regarding depressive symptoms were more consistent than those for anxiety. Sudanese refugee parents reported higher levels of depression symptoms of their children in comparison to Ugandan children who did not reside in camps (Paardekooper, de Jong, & Hermanns, 1999). Ekblad (1993) also found that children in Swedish refugee camps had higher reports of depression and that 45% of youth reported that these symptoms had increased since arrival at the Swedish refugee camp. In a sample of 403 Palestinian children, the most frequently endorsed depression items were crying (25.6%), lack of energy (22.1%), restlessness (18.4%), and feeling lonely (18.4%; Thabet et al., 2004). Additionally, Miller (1996) found a moderately strong relation between parent report of youth depressive symptomatology and self-report of mothers’ physical and psychological health status (r = .34), with this relation being stronger for girls than for boys.
Aggression
Few studies in this review focused on externalizing behaviors of youth in refugee camps. Paardekooper and colleagues (1999) found that Sudanese refugees had higher levels of aggression than Ugandan nonrefugee camp children. Additionally, Irshad and Bano (2004) assessed aggression levels of 150 Afghan children in the Shamshatoo refugee camps. Aggression was measured using the human figure drawing test. Results indicated that 36% of the children drew overly extended arms, and the authors concluded that this was an emotional indicator for having aggressive tendencies. The authors concluded that the drawing analysis suggested that male children were more aggressive than female children because males were significantly more likely to draw emotional indicators of aggression (long arms, straight lines, weapons, heavy shading, etc.). Six percent of youth drew weapons and knives, which was interpreted as an emotional indicator for aggression.
Somatic symptoms
In the current review, seven studies examined youth somatic symptoms. Parent reports of youth somatic symptoms indicated that Guatemalan Mayan youth residing in two refugee camps in the Mexican state of Chiapas most commonly endorsed headaches (M = 0.91) and stomachaches (M = 0.78; Miller, 1996). Not only are youth in refugee camps experiencing high levels of somatic symptoms, but Ekblad (1993) found that these symptoms increased after arrival into the camp for 41% of youth in Swedish refugee camps. Further, Paardekooper and colleagues (1999) reported that parents of Sudanese refugee camp children reported that their children experience more headaches and their psychosomatic complaint scores were higher than the nonrefugee Ugandan youth. Finally, Muecke and Sassi (1992) compared results from a sample of Cambodian adolescents in a Thailand refugee camp to Thai youth who had resettled in the United States. Both the refugee youth and resettled youth most frequently reported experiencing headaches (M = 1.99, M = 2.27, respectively), fainting, dizziness, or weakness (M = 1.32, M = 1.88, respectively), restlessness (M = 1.23, M = 1.69, respectively), and feeling tense (M = 1.2, M = 1.85, respectively). Across each of the symptoms, the resettled group endorsed more somatic symptoms than the refugee youth. Muecke and Sassi attributed these unexpected findings to the older age and thus possible developmental maturity of the resettled group (i.e., refugee age = 14.1 years old, resettled youth age = 17.5 years old).
Local syndromes
Two studies focused on various locally defined mental health disorders. For example, Smith, Sabin, Berlin, and Nackerud (2009) found that caregiver reports on a youth health questionnaire indicated that 48.4% of youth in refugee camps in Chiapas, Mexico, have experienced susto, a fright condition associated with symptoms of PTSD, anxiety, and depression. Additionally, Betancourt, Speelman, Onyango, and Bolton (2009) conducted a qualitative study focused on locally defined mental health problems for Acholi youth in Ugandan refugee camps. Authors identified five commonly reported mental health symptoms in youth: (1) Two tam, which is described as similar to dysthymia, (2) Kumu, which is described as being similar to major depressive disorder, (3) Par, which is also similar to dysthymia, (4) Ma Lwor, which is comparable to generalized anxiety disorder, and (5) Kwo Maraco, which shares some symptoms with oppositional defiant disorder.
Qualitative Studies
Qualitative studies provided additional details about experiences in the refugee camps and the implications of potentially traumatic events. Four of the studies also included a focus on qualitatively reporting the mental health outcomes experienced by youth (Betancourt et al., 2009; Miller, 1996; Rothe, 2005; Smith et al., 2009). For instance, Rothe (2005) qualitatively assessed the experience of fleeing their country of origin, life in the refugee camp, and how past events contributed to mental health outcomes. Results indicated that crossing the ocean from Cuba to America and living in refugee camps were the two most traumatic experiences for the youth, which reportedly caused youth to avoid places that reminded them of the trauma, such as beaches. These events also caused reoccurring nightmares of dead relatives and human corpses being eaten by sharks. Youth reported that the crowdedness, noise, and mosquitoes in the camp caused them to be irritable and act out aggressively. Adolescent’s particularly felt that camp life was boring, repetitive, and reduced their autonomy, causing adolescent boys to act out by escaping from the camps and girls to act out sexually. Miller (1996) also obtained qualitative reports on the refugee youth experience. Results indicated that youths’ knowledge about leaving Guatemala was obtained from their parents, and yet many of the children talked about the violence occurring in Guatemala, experienced nightmares, and were afraid to return home. Betancourt et al. (2009) and Smith et al. (2009) both conducted qualitative surveys to get a better understanding of the local mental health symptoms that were affecting children in the refugee communities. Both of these studies are discussed in further detail in the previous section. Smith et al. (2009) found that when children experienced local mental health symptoms, doctors and locally traditional methods of treatment were sought out. Limpias described as a spiritual cleansing was reported by parents as a very effective method to treat children. Betancourt et al. (2009) reported that when children experienced local mental health symptoms, the common method of treatment was to sit with children, provide comfort, and help the child talk about their problems.
Mental Health Outcomes: Second-Order Themes
Mental health outcomes by continent
Seventeen of the studies included prevalence rates of the mental health outcomes. For studies conducted in Asia, 87% of Iraqi youth had PTSD (Ahmad, Sofi, Sundelin-Wahlsten, & von Knorring, 2000), youth aggression rates reportedly ranged from 11.5% (Mollica et al., 1997) to 45.3% (Irshad & Bano, 2004), and youths’ depression and anxiety symptoms ranged from 1.4% to 11% (Mollica et al., 1997; Muecke & Sassi, 1992; Yurtbay, Alyanak, Abali, Kaynak, & Durukan, 2003). Finally, Izutsu et al. (2005) reported that 5–70% of Pakistani youth endorsed elevated rates of suicidal ideation across the different camps located in Asia.
Studies from Europe indicate that 94% of youth exhibited symptoms of PTSD (Goldstein, Wampler, & Wise, 1997) and 9.7% of youth had elevated aggression in Croatian refugee camps (Ajdukovic & Ajdukovic, 1993). Four studies examined anxiety and depression of youth in refugee camps in Europe and prevalence rates ranged from 9.1% to 95.5% (i.e., Ajdukovic & Ajdukovic, 1998; Ekblad, 1993). Finally, one European study found that 37.7% of youth residing in Bosnian refugee camps endorsed suicidality (Goldstein et al., 1997).
Two studies in Africa examined the prevalence of PTSD for youth in camps; Peltzer (1999) reported a prevalence rate of 20%, whereas Morgos, Worden, and Gupta (2008) found that 75% of youth endorse PTSD symptoms. Additionally, 5.3% of youth in Ugandan refugee camps had higher rates of aggression tendencies (Paardekooper et al., 1999). The prevalence of anxious and depressive symptoms across youth in Africa ranged from 3.8% to 38% (Morgos, Worden, & Gupta, 2008; Paardekooper et al., 1999). Paardekooper et al. (1999) also examined the prevalence rate of suicidal thoughts and found that 4.4% of youth participants experience symptoms.
In North America, the prevalence rates of PTSD ranged from 0% to 88% for youth in refugee camps (Miller, 1996; Rothe, 2005; Smith et al., 2009). Rothe (2005) also measured the prevalence of aggressive behavior and found that 73% of youth in Guantanamo refugee camps exhibit elevated aggression. Two studies from North American camps examine the prevalence of anxious and depressive symptoms. Rothe (2005) found that 35% of youth experience generalized anxiety and depressive feelings, whereas Smith et al. (2009) found that 48.4% of children experienced symptoms of anxiety and depression. Rothe (2005) examined the rate of youth experiencing suicidal thoughts in North American camps; results indicated that 25% of girls and 14% of boys between the ages of 7 and 12 years old endorsed suicidal thoughts.
Mental health outcomes by duration in camps
The duration of time youth spent living in camps varied greatly across studies from 4 weeks (Ahmad et al., 2000) to 5 or 6 years (Van Ommeren et al., 2001). Nine studies did not report how long their sample had lived in the camps. Although no consistent pattern of symptomatology was seen across these variations in duration, some researchers examined the impact of the duration within their study. Specifically, Ahmad, Sofi, Sundelin-Wahlsten, and von Knorring’ (2000) results suggest that the length of time living in camps was associated with more severe PTSD symptoms. In contrast, Ajdukovic and Ajdukovic (1998) reported that youth who lived in camps longer had fewer PTSD symptoms than those who lived in camps for a shorter period of time. Additionally, this inverse relation appeared to be stronger for males than females (Stien et al., 1999).
Mental health outcomes by trauma experience
Fifteen studies in this review reported on level or types of trauma experienced by youth and the prevalence rates of mental health outcomes. Across these studies, there was a wide range of types, for instance, Muecke and Sassi (1992) reported that the majority of youth experienced few to no traumatic events, and others reported youth were exposed to a wide range of traumas (i.e., Morgos et al., 2008). Studies in which the majority of youth personally witnessed the death or torture of another human (usually a family member) had the highest prevalence rates for mental health outcomes (Ahmad et al., 2000; Goldstein et al., 1997; Morgos et al., 2008; Rothe, 2005). For example, Goldstein and colleagues (1997) reported that a majority of youth (58.9%) witnessed the injury or death of a parent or sibling, and rates of distress were significant for this sample of youth, with 94% meeting criteria for PTSD, 95.5% had high levels of anxiety, and 37.7% of youth felt life was not worth living. Furthermore, Rothe (2005) assessed mental health symptoms of Cuban youth in Guantanamo camps that had witnessed people drowning and saw human corpses devoured by sharks; 88% of the sample endorsed PTSD symptoms. Only one study in which the majority of youth witnessed the death of a family member (63%) found comparatively low levels of mental health outcomes; only 16.1% of youth reported experiencing survival guilt, 12.4% of youth engaged in risk taking behaviors, 4.4% of youth experienced suicidal thoughts, and 3.8% of youth were often very sad (Paardekooper et al., 1999).
Studies that reported separation from one or more family members as a main traumatic experience had comparatively lower mental health outcome prevalence rates. Additionally, studies that reported few to no traumatic events showed youth having lower rates of negative mental health outcomes than studies, where multiple traumatic events were reported by youth (i.e., Miller, 1996; Yurtbay et al., 2003). For instance, Ajdukovic and Ajdukovic (1993) studied youth in Croatian refugee camps and reported that 46.2% of youth in the study were separated from one or more family members and experienced no other traumatic event after coming to the refugee camp. Among this group, 9.1% were reportedly fearful, 17.2% experienced separation fears, 9.7% exhibited signs of aggression, and 16% had an eating disorder. Additionally, 24.4% of youth exhibited no negative mental health symptoms, 24.4% exhibited 1–2 symptoms, 18.4% exhibited 3–4 symptoms, and 16.5% exhibited 5–6 symptoms. Miller (1996) conducted a study in a Mexican refugee camp and found that youth recalled few if any memories of direct violence or trauma due to their refugee experience and 0% of youth report PTSD symptomatology.
Mental health outcomes by age and gender
Eight studies in this review examined age as a possible contributing factor to mental health outcomes. Three studies found that older youth had worse mental health outcomes than younger youth (Ahmad et al., 2000; Ajdukovic & Ajdukovic 1998; Muecke & Sassi, 1992). However, Yurtbay et al. (2003) found that 9- to 10-year-old children had the highest levels of depression compared to any other age-group. Morgos and colleagues (2008) studied youth ranging from 6 to 17 years old and found that older children (13–17 years old) were exposed to more war trauma experiences and therefore reported greater levels of somatic symptoms, were more likely to relive the trauma, felt more isolated, and had more grief symptoms than younger aged youth. Furthermore, Rothe (2005) found age and gender differences. Specifically, females between 7 and 12 years old experienced the highest rate of severe PTSD symptoms, adolescents (13–19 years) had the lowest PTSD scores, and adolescent boys were the most likely to experience suicidal thoughts. Two studies found that age did not have an effect on mental health outcomes (Miller, 1996; Stein et al., 1999).
Mother’s mental health as a protective factor
Four studies in this review also examined how parental mental health may impact youth symptomology. Results suggest that the mother’s coping ability and mental health symptoms were correlated with their child’s level of mental health symptoms (Ajdukovic & Ajdukovic, 1993, 1998; Ekblad, 1993; Miller, 1996). For instance, Adjukovic and Ajdukovic (1993) found that youth’s level of stress reactions were correlated with their mother’s ability to cope with displacement such as their general dissatisfaction with their own situation, having a poor relationship toward own children, and having a poor relationship with others housed together. Ekblad (1993) found that 59% of mothers felt they coped well in the refugee camps, and mothers who said they coped well had children who experienced few negative mental health outcomes.
Discussion
The purpose of the present review was to systematically assess the research on mental health outcomes for youth living in refugee camps. Given that a majority of displaced youth often live in a refugee camp for extended periods of time and can be exposed to a range of traumatic experiences, it is important to understand the mental health of these youth while they are living in camps. Although reviews have provided summaries of adjustment postcamp life (Lustig et al., 2004; Reed et al., 2012), to date, no review has provided a clear picture of what is known about mental health functioning during refugee camp life or the methods used to capture mental health for what is often a multinational, culturally varied population. Although the measures and research designs across studies varied considerably, the results of the present review suggest that many youth in refugee camps (although not all youth) can and do have significant mental health issues and that the maladjustment reported often ranges across the pathology continuum.
Variations Across Studies
Perhaps the most challenging issue for establishing the prevalence and nature of mental health for youth in refugee camps is the wide variety of procedures used across studies. First, studies varied greatly in the reporter utilized to assess youths’ mental health. Parent report, youth report, and behavioral observations are all ways that youth mental health status was assessed. Although it is often considered a strength to have a multi-informant approach to assessing mental health outcomes for youth, the methods were not consistent across studies in terms of reporter, making comparisons difficult. For example, for somatization, four studies used parent reporter, two studies used youth self-report, and one study used both parent and child report. The same pattern can be seen with externalizing behaviors (i.e., Irshad & Bano, 2004; Rothe, 2005). Given that some research suggests that youth tend to be better reporters of their internalizing problems and parents tend to be better reporters of youth’s externalizing problems (De Los Reyes & Kazdin, 2005), it is important to consider the source of the information provided. Sometimes including multiple reporters means that the information will, by definition, be different; however, only eight studies used multi-informants to measure the mental health construct of interest, and in those studies, the findings across reporters tended to be consistent.
Second, studies varied in the measures used to gather information on mental health outcomes. Measures ranged from commonly used, psychometrically sound tools, to scales that were developed by authors or qualitative interviews. This variety of measures used can be problematic because it does not allow for comparison across studies that measure the same mental health outcomes and could suggest some problems with construct validity in the data. The accuracy of the data gathered from these methods may be questionable and their applicability to samples with differing circumstances and cultures from the Western world (where many of the measures were created) provides some caution in interpretation of the findings.
Third, there was a lack of consistency regarding how measures were translated. Just under half of the studies (eight) did not describe procedures for their translation methods. Without knowing the rigor of the translation, it is not possible to know how valid the translated measures were or if the participants understood the meaning of the questions. For example, terms such as mental health problems, depression, and anxiety may not be culturally common and careful translation of methods are needed to ensure respondents understand the nature of what the question is asking. Of the studies that did describe translation methods, 30% relied solely on a one-way/direct translation method and only one study used key informant translation and pretests. This is unfortunate given that key informant translations and pretests are important for checking the applicability of measures in a given sample, as it is possible that some local populations do not have words or the same conceptual understanding of many common psychological terms (Peltzer, 1999).
Fourth, the studies reviewed examined youth living in camps across the world, making generalizations of results difficult, as conditions can vary greatly within a country and even more so across different countries and resources (Izutsu et al., 2005). Additionally, the duration of time youth lived in refugee camps differed across studies, and several articles did not report on this variable. For the studies that did collect data on the duration of time living in a camp, this appeared to be an important variable that impacted severity of mental health outcomes (Ahmad et al., 2000; Izutsu et al., 2005).
Prevalence of and Nature of Mental Health Outcomes
Although this current review is the first of its kind by focusing on mental health outcomes of youth in refugee camps, the results of this review still had many similarities to other reviews on displaced youth and adults. Among the reviews discussed previously in Introduction section, the prevalence of mental health disorders varied widely across studies, and this present review found the same results in this regard (Fazel et al., 2012; Lustig et al., 2004; Reed et al., 2012). The greatest variability across studies was seen for youths’ anxiety and PTSD symptoms. For instance, Miller (1996) found no significant levels of PTSD symptoms, indicating possibly no unique contribution of camp life effects on youth mental health. Rothe (2005), however, found high levels of PTSD, with 57% of youth meeting diagnostic criteria. One possible explanation for these differences in results is that Miller’s (1996) sample experienced no direct exposure to war violence, whereas many of those in the Rothe’s (2005) sample experienced war violence. Also, consistent with Fazel et al.’s (2012) previous review of adults, the rate of PTSD among refugee youth appears to be higher than the age-matched Western populations. For example, in one study of similarly aged youth in the United States, there was an average prevalence of 8.8% of PTSD symptoms for youth who experienced trauma and a prevalence of 18.8% for traumatized youth in foster care (Salazar, Keller, Gowen, & Courtney, 2013). These rates have a narrower range than the PTSD prevalence range of 0–87% as found in the current review. The wide range of results could be due to different refugee camps having widely different resources, levels of violence in the camp, and different trauma experience prior to camp life. Furthermore, some studies defined PTSD symptoms as an indicator of PTSD and others used PTSD diagnosis as an indicator of PTSD (Ahmad et al., 2000; Miller, 1996).
Similar discrepancies across studies were found for anxiety. For example, although Muecke and Sassi (1992) reported lower levels of anxiety for youth in refugee camps than for resettled youth, Yurtbay et al. (2003) found higher levels and a higher prevalence of anxiety for youth in refugee camps compared to nonrefugee youth living in a similar area. Muecke and Sassi (1992) attribute their surprising results to refugee youth having higher levels of hope than the comparison group of youth who had resettled. Additionally, the prevalence of anxiety reported in the current review is higher than that of similarly aged Western youth, typically around 25.1% (Merikangas et al., 2010).
The prevalence of depressive symptoms ranged from 35% (Morgos et al., 2008) to 90% (Goldstein et al., 1997) across studies. The average prevalence rate in similarly aged youth in the United States is 11.4% (Center for Behavioral Health Statistics and Quality, 2015), which again suggests that refugee camp life or specific-related traumatic experiences may be uniquely associated with high levels of symptomatology.
Although few studies examined the prevalence of aggression or conduct problems among youth in refugee camps, those who did found the prevalence to be around 36% and that boys displayed more aggression than girls (Irshad & Bano, 2004) and that the rate was higher and more severe than a nonrefugee comparison group (Paardekooper et al., 1999). Although this gender difference in aggression is also found in Western populations, the current review shows that refugee youth may have a higher prevalence of aggression than Western youth, which is typically 5–7% (Schlack & Petermann, 2013).
For somatic symptoms, across studies, reporters, age of youth, and location of refugee camps, the most common somatic symptom of youth was headaches. However, qualitative research also highlighted the high prevalence of nightmares among youth (Rothe, 2005). Ekblad (1993) found 41% of refugee youth present with elevated somatic symptoms, which should be interpreted along with Paardekooper et al.’s (1999) finding that when compared to nonrefugee youth, refugee youth reported more somatic symptoms. Again, Western youth appear to have a lower prevalence of somatic symptoms, ranging from 3.5% to 8.3%, than refugee youth in the current review (Schlack & Petermann, 2013).
Context and Mental Health Concerns
Consistent with Fazel and colleagues’ (2012) review, who found that adult refugees from different countries of origin have different types and duration of exposure to traumatic events, cultural differences in handling distressing events, and different patterns of psychological distress, the current review found the prevalence of youth’s mental health symptoms to vary based on the location of and duration of time spent in refugee camps. Specifically, mental health symptoms across and within continents were found to vary widely, and although it appears as if location of camp may be an important contributor of mental health outcomes, there was no clear and distinct pattern of results, contrary to Frazel et al.’s (2012) review. A clear pattern of results may have not been found for location of camps due to a number of reasons. One reason could be, unlike Fazel et al.’s (2012) review, which focused on high-income resettlement countries, availability of resources was not taken into account in this review. These resources could have varied widely within camps in the same continent, contributing to greater variability among camps within the same continent making cross-continental comparisons challenging. Similarly, the amount of war exposure to violence could have varied widely within camps located in the same continent.
Inconsistencies were found for associations between symptoms and the duration of time spent living in camps. Specifically, some results suggest that the duration of time living in camps was associated with more severe PTSD symptoms (Ahmad et al., 2000), whereas other results suggest that duration of time spent in camps was associated with fewer PTSD symptoms and feeling happier (Izutsu et al., 2005). These differences may be partially explained by different conditions of camps. As Toole’s (1994) analysis points out, conditions and availability of resources can vary greatly across camps; however, the specific details of what youth in a given camp experienced are not provided in most published studies, again, making comparisons from the data at one camp to another challenging.
Consistent with previous reviews of adults and displaced youth, exposure to great amounts of war violence, stress, or traumatic experiences was associated with more severe PTSD and depressive symptoms (Ajdukovic & Ajdukovic, 1998; Mollica et al., 1997; Morgos et al., 2008; Paardekooper et al., 1999; Rothe, 2005; Thabet et al., 2004). There was also a wide range of traumatic experiences reported across studies, but it appeared that witnessing death or torture was consistently associated with worse mental health outcomes.
Age and gender differences in symptoms were not largely examined among studies. However, the majority of studies that examined age found that older youth had worse symptomatology than younger youth (Ahmad et al., 2000; Ajdukovic & Ajdukovic 1998; Muecke & Sassi, 1992). This could be because older youth tended to experience more and different kinds of traumatic events (Miller, 1996). With this being said, some studies found no effect of age differences (Miller, 1996; Stein et al., 1999). Furthermore, there were no studies of symptoms in early childhood indicating a need for future studies with this age-group. Inconsistencies may be explained by few studies examining age and among the studies that did children’s trauma experiences varied across studies (Ahmad et al., 2000; Muecke & Sassi, 1992).
For gender differences, consistent with Western studies of aggression, boys displayed more aggression than girls (Card, Stucky, Sawalani, & Little, 2008; Irshad & Bano, 2004), Western studies also commonly find that girls experience more internalizing symptoms, such as depression, than boys (Nolen-Hoeksema, 2001), which was not found among the populations covered in this review. These results suggest the refugee youth may be equally susceptible to internalizing problems, regardless of gender, perhaps due to similar experiences of trauma or displacement. However, across nonclinical Western samples and refugee youth, boys tend to show more aggression than girls, which may be partially explained by biological/hormonal differences (Card et al., 2008).
Specific to refugee youth, evidence also stressed the importance of maternal mental health and coping with stressful events, as this is associated with youth’s PTSD, anxiety, depression, and somatic symptoms (Ajdukovic & Ajdukovic, 1998; Miller, 1996). This finding is consistent with prior reviews, which also identified parental well being as a protective factor against mental health symptomology for displaced youth (Fazel et al., 2012; Reed et al., 2012).
Although not surprising given the high level of exposure to what are usually nonnormative events (i.e., war), youth in camp living appear to have high levels of mental health problems. While this trend was consistent across studies, not all of the results were consistent with this conclusion, thus the need for additional and more methodologically rigorous study of youth adjustment to camp living.
Limitations
Although the present review identifies several important findings, it is not without limitations. For example, the present review excluded literature on refugee youth who live in noncamp settings, such as dispersed communities. The review also did not distinguish between different refugee camp conditions and the resulting mental health outcomes. Perhaps knowing the current conditions in the area around the camp or the general economic stability of a given country could contribute to conditions in the camp and the youths’ general mental health status. Additionally, only studies that were available in English were included in the current review, excluding two studies from the review, which only included abstracts in English, thereby, potentially limiting the amount of information gained.
Directions for Future Research
To improve the validity of measurement of mental health, it is important that researchers not only begin to use similar or consistent measurement tools so that the findings can be generalized across studies but it is also important for researchers to utilize local or even newly developed measurements along with standardized measures to ensure that outcomes are being measured in culturally appropriate ways. This issue becomes particularly challenging for studying refugee populations because studies can involve many different languages, countries, and cultures. As the majority of these measures are created in the West, it may not be applicable or even translatable for every study in different areas to use the same set of measures (e.g., some countries have very different conceptualizations of mental health). Therefore, it may be beneficial for studies to use a combination of both common standardized measures and locally created measures or new measures designed for a specific culture or language group. For example, researchers could use checklists of observable symptom behavior using local terms along with their standardized measurement in order to see if the checklist was consistent with what was being reported on the standardized measure. Researchers may also need to report their translation methods in a more transparent and detailed manner in their studies, so readers can know exactly what methods were used. It is also important that researchers include back-translation methods and pretest/discussion with local key informants in research projects, so that measurement tools, items, and psychological constructs can be carefully reviewed for construct validity and applicability to the cultural group represented by the sample in the study.
Future research should continue to incorporate comparison groups. For example, among the studies reviewed, comparison groups included youth from surrounding areas who are not in refugee camps, already resettled children as the comparison group, and displaced youth from the same area who are not in camps (Muecke & Sassi, 1992; Paardekooper et al., 1999; Yurtbay et al., 2003). Although these are all good methods of acquiring comparison groups, using displaced youth who do not reside in camps will allow examination of the possible unique mental health outcomes from camp life, as both groups experience displacement. Finally, multiple assessments during camp life are also important, particularly for refugee youth who are experiencing long-term stay in the camps. Having multiple time points of data would allow for better assessment of the long-term effects of camp life, as mental health symptomology change over time.
Implications for Practice, Policy, and Research
Significant variation exists in outcomes of refugee youth with regard to type of mental health problem investigated and levels of mental health problems reported. This review may inform policy makers and practitioners of the importance of mental health services and intervention programs in refugee camps for youth.
Given qualitative results of youth’s reported boredom and lack of autonomy, intervention strategies within camps can work to increase recreational activities for youth and to promote feelings of control and independence.
Youth intervention strategies should consider the important role that parental health and coping skills can play in refugee youth’s mental health.
Reviews should examine the impact of refugee camp conditions (i.e., socioeconomic status of country, war zones, overcrowding, etc.), mental health outcomes from refugee camp life, and mental health levels of refugee youth who live in noncamp settings.
Future research should examine the impact of age and developmental stages on mental health outcomes and should also include comparison groups, clearly described translation methods, and the use of consistent measurement tools.
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) received no financial support for the research, authorship, and/or publication of this article.