Explain the psychological issues that may result from your topic.
- Describe the most effective assessment measure that could be used, and explain why you selected this.
- Explain the treatment options available for children and adolescents involved with your selected disorder.
- Explain how culture may influence treatment.
- 3 paragraphs
Children and Youth Services Review 73 (2017) 93–99
Contents lists available at ScienceDirect
Children and Youth Services Review
j ourna l homepage: www.e lsev ie r .com/ locate /ch i ldyouth
Unaccompanied migrant children in the United States: Predictors of placement stability in long term foster care
Thomas M. Crea a,⁎, Anayeli Lopez a, Theresa Taylor b, Dawnya Underwood b a Boston College School of Social Work, 140 Commonwealth Ave., Chestnut Hill, MA 02467, United States b Lutheran Immigration and Refugee Service (LIRS), 700 Light Street, Baltimore, MD 21230, United States
⁎ Corresponding author. E-mail address: email@example.com (T.M. Crea).
http://dx.doi.org/10.1016/j.childyouth.2016.12.009 0190-7409/© 2016 Elsevier Ltd. All rights reserved.
a b s t r a c t
a r t i c l e i n f oArticle history: Received 11 August 2016 Received in revised form 9 December 2016 Accepted 9 December 2016 Available online 11 December 2016
Tens of thousands of unaccompanied children have migrated to the United States in recent years, particularly from Central America, but also from a range of other countries. These children are fleeing unprecedented levels of violence in their home countries, and often are seeking reunification with parents and family members (i.e., sponsors). Some children are not able to live with a sponsor, and these children are placed in federally funded foster care. Yet, virtually nothing is known about how these children fare during placement. Literature related to domestic child welfare has demonstrated the importance of placement stability to children’s future well- being. Using an exploratory design, the purpose of this study is to examine the placement stability of unaccom- panied youth while in long term foster care from 2012 to 2015, and how pre-migration, transit, and post-place- ment risk factors are each associated with placement changes for these children (n = 256). Results show that experiencing violence in home countries, and significantly acting outwhile in care, were associatedwith a higher likelihood of changing placements. Migration-related trauma was not significant, but fear of returning to home countries, and suffering trauma unrelated to migration, each was associated with a lower likelihood of changing placements. Children fromNorthern Triangle countries (El Salvador, Guatemala, andHonduras)weremore likely to have experienced a failed family reunification prior to entering foster care. These results are discussed in light of the need to adopt a global perspective in child welfare that interprets children’s behavior in the larger context of pre-migration experiences and culture.
© 2016 Elsevier Ltd. All rights reserved.
Keywords: Unaccompanied children Migrant youth Placement change Immigration Foster care Global child protection
Increasing numbers of unaccompanied children are entering the United States through the Southern border, particularly originating from the Northern Triangle countries of El Salvador, Guatemala, and Honduras (U.S. Customs and Border Protection [CBP], 2016). This in- crease began in 2011 (Women’s Refugee Commission, 2012), with 24,403 children arriving in fiscal year (FY) 2012, and 38,759 in FY2013 (CBP, 2016). A significant surge of 68,541 children in FY2014 overwhelmed government facilities and placed the issue under in- creased media attention (American Immigration Council, 2015; Chishti & Hipsman, 2015). Although these numbers dropped to 39,970 in FY2015, this pattern of migration has since continued, with the num- bers for FY2016 (59,692) approaching the levels of FY2014 (CBP, 2016). Most unaccompanied children are taken into custody by the Office of Refugee Resettlement (ORR) and eventually placed with an adult spon- sor, usually a relative. For a small number of children and youth, ranging between 5.0% (Roth & Grace, 2015) and 35.0% (Byrne & Miller, 2012), sponsors are unavailable or deemed unsuitable for placement. In these
cases, children are placed in long term foster care (LTFC) while they await deportation and legal status hearings (Byrne & Miller, 2012).
Placement stability while in foster care is important for children’s well-being. Research on domestic child welfare services has consistent- ly demonstrated a strong association between frequent placement moves in foster care and poor outcomes for foster care children (Barth & Jonson-Reid, 2000; Newton, Litrownik, & Landsverk, 2000; Rubin, O’Reilly, Luan, & Localio, 2007). However, virtually nothing is known about placement stability for unaccompanied children in LTFC or the ex- planatory factors of stability. Unaccompanied children enter LTFC for reasons related largely tomigration, while children in domestic systems enter foster care for reasons mostly related to maltreatment. Therefore, these two systems, and the experiences of children within each system, are likely to be very different. Existing literature suggests that many un- accompanied children may have significant risk factors such as trauma from experiencing violence prior to and during the journey to the U.S., and ambiguous legal status (Byrne & Miller, 2012), all of whichmay in- fluence stability in foster care.
Using a “stages ofmigration” framework, Pine andDrachman (2005) outline key variables to be considered when applying child welfare practice principles to immigrant children and families (p. 544). These stages include the pre-migration or departure stage; the transit or
Table 1 Bivariate comparisons – Northern Triangle (vs. other countries of origin).
Northern Triangle (n = 190)
Other (n = 66)
M(SD) or % M(SD) or %
Age (years)⁎ 16.24 (1.62) 15.24 (3.57) Gender (1 = female) 26.18 19.70 Abandoned in home country 27.23 24.24 Experienced violence in home country
Failed family reunification⁎⁎ 34.03 13.64 Fear of returning to home country⁎ 15.71 28.79 Trauma (not related to journey) 23.56 28.79 Trauma (related to journey) 6.81 12.12 Significant acting out 4.71 7.58
⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.
94 T.M. Crea et al. / Children and Youth Services Review 73 (2017) 93–99
intermediate stage; the resettlement stage; and the return to country of origin stage. Unaccompanied youth in LTFC find themselves in a formal social service system, yet lacking legal status that would qualify them for additional supports. Following Pine and Drachman’s (2005) frame- work, these youth are therefore in the resettlement stage, prior to a de- termination of legal permanency in the US or a potential return to their country of origin.
The purpose of this study is to examine theplacement stability of un- accompanied youth while in long term foster care from 2012 to 2015. Guided in part by the “stages of migration” framework (Pine & Drachman, 2005), this study examines how pre-migration and transit risk factors, and risk factors that emerge after placement, are each asso- ciated with children’s likelihood of experiencing a placement change in care. Given the recent surge of migration from Central America, the study also focuses on the extent to which children’s experiences differ based on whether the migrated from the Northern Triangle countries of Central America, compared to those from other countries.
2. Pre-migration and transit risk factors
Unaccompanied children come to the United States from all over the world, and have done so for decades (Roth & Grace, 2015). Increasingly, however, the majority of unaccompanied children in 2015 originated from the Northern Triangle countries of Guatemala (25%), El Salvador (24%), and Honduras (27%), where violence, homicide and poverty rates are strikingly high, as well as from Mexico (23%) (American Immigration Council, 2015). A study of unaccompanied children apprehended at the U.S.-Mexico border, conducted by the United Na- tions High Commissioner for Refugees (UNHCR, 2014), determined that over half of the children interviewed (n = 404) suffered forcible displacement because of events that warranted, or couldwarrant, inter- national protection. Nearly half (48%) of the children in the study were displaced because of violence caused by drug cartels and gangs, or by government actors. Twenty-one percent of the children reportedhaving been victims of abuse and violence in their homes. Among the 102Mex- ican children interviewed (25.3% of the sample), 38% revealed they were recruited into and exploited by the criminal industry of human smuggling (UNHCR, 2014).
In addition to family maltreatment, gang violence, human traffick- ing, and rape experienced by many youth in their home countries (UNHCR, 2014), the migration journey itself may be a source of addi- tional trauma (Pine & Drachman, 2005). Unaccompanied youth cross the border through variousmeans includingbybus, on the roof of trains, hiding in trucks, or on foot, often witnessing violence, rape, and death along theway (Griffin, Son, & Shapleigh, 2014). These forms of extreme adversity can lead to a multitude of mental health problems that are correlated with the number of traumatic events experienced by youth (Almqvist & Broberg, 1999; Derluyn, Mels, & Broekaert, 2009; Kirmayer et al., 2011; Sourander, 1998). Longitudinal studies have also shown a link between refugee minors’ exposure to traumatic events and their long term recovery. The related psychological distress is both severe and chronic during youths’ initial one-two years in the host country, such that recovery is a long process (Almqvist & Broberg, 1999; Bean, Eurelings-Bontekoe, & Spinhoven, 2007; Jensen, Skårdalsmo, & Fjermestad, 2014; Keles, Friborg, Idsøe, Sirin, & Oppedal, 2015). Given the violence and other traumatizing experiences often endured by unaccompanied children before and during their mi- gration journey, their recovery process may be complex and difficult upon their arrival to the U.S. – and very little research or practice guid- ance exists in the literature pertaining to the unique needs of this pop- ulation of children.
3. Placements for unaccompanied children
In the U.S., once an unaccompanied child is apprehended at the bor- der, he or she is screened by a Customs and Border Protection (CBP)
officer to determine if he or she is eligible for temporary legal protection (American Immigration Council, 2015). Some have criticized this proce- dure, expressing that CBP is not the appropriate agency to assess chil- dren for signs of trauma, abuse or persecution (American Immigration Council, 2015), and children may be reticent to disclose these sensitive issues to screening officers (Roth & Grace, 2015). In addition, the pro- portion of younger children migrating to the U.S. has increased over time (ORR, 2016b). While the numbers of children aged 15–16 have remained fairly constant (between 36%–40%), those aged 0–12 accounted for 11% of the population in FY2011 and FY2012, but doubled to 22% in FY2014. Similarly, the proportion of children aged 17 has de- creased to a low of 30% in FY2015 from a high of 40% in FY2011 (ORR, 2016b).
Amajority of unaccompanied children are referred by CBP to the Of- fice of Refugee Resettlement (ORR) and most are then placed with an adult sponsor while they await Immigration Court hearings (Chishti & Hipsman, 2015; Roth & Grace, 2015). In FY2014, these numbers totaled 52,515 children; this number dropped in FY2015 to 27,840, but in- creased in FY2016 to 52,147 (ORR, 2016c). Children are discharged to sponsors (usually family members) across multiple states. In FY2016, the most common states included the following: California (14.2%); Texas (12.6%); Florida (10.1%); Maryland (7.4%); Virginia (7.2%); New Jersey (5.1%); Georgia (3.3%); Massachusetts (3.0%); North Carolina (2.9%); and Tennessee (2.6%) (ORR, 2016c).
Children that are not able to find a viable sponsor in the U.S. may be placed in long term foster care placements or group homes (ORR, 2015). Public data are not available related to the number and types of these placements. Whether placed with a sponsor, or placed in LTFC, children remain in the custody of ORR until a decision is made on their legal sta- tus, or until they reach age 18 at which point they face the possibility of adult detention and/or immediate deportation. These children therefore find themselves in legal limbo while in LTFC.
ORR defines LTFC as community-based foster care, where eligible children are transferred if it is determined that they will remain in ORR custody for a significant length of time (ORR, 2016a). For children in LTFC, the goal of ORR is to place unaccompanied children in the least restrictive setting, while helping them reunify with their parents or other appropriate caregivers. ORR contracts with state-licensed ORR-funded care providers (ACF, 2012) such as Lutheran Immigration Refugee Service (LIRS) and the United States Conference of Catholic Bishops (USCCB), who then subcontract with agencies that license and oversee foster placements in local jurisdictions. When possible, these youth are placedwith foster families rather than group-based or institu- tional care (Carlson, Cacciatore, & Klimek, 2012). Studies with refugee children in other countries show better outcomes for children in foster care and have recommended foster care over group-based placements (Duerr, Posner, & Gilbert, 2003) and the same is true for children in
95T.M. Crea et al. / Children and Youth Services Review 73 (2017) 93–99
the domestic child welfare program in the United States (Barth, 2002). These providers also deliver services for the youth in care including ac- cess to legal services, education, culture, language and religious obser- vation, physical and mental health care, and recreation activities.
To provide care for unaccompaniedmigrant children, ORR has in es- sence created a stand-alone child welfare system that operates parallel to domestic systems. In domestic systems, home study assessments are required of all foster parents prior to receiving a child to assess their family’s suitability as a foster care placement (Crea, Barth, & Chintapalli, 2007). Unlike the domestic child welfare system however, the Trafficking Victims Protection Reauthorization Act (TVPRA) of 2008 only requires home studies for the adult sponsors under the fol- lowing circumstances: (1) The child is a victim of a severe form of traf- ficking in persons; (2) the child is a special needs child with a disability as defined by section 3 of the Americans with Disabilities Act of 1990 (42 U.S.C. 12102(2)); (3) the child has been a victim of physical or sex- ual abuse under circumstances that indicate that the child’s health or welfare has been significantly harmed or threatened; or (4) the child’s sponsor clearly presents a risk of abuse, maltreatment, exploitation or trafficking, to the child based on all available objective evidence (ORR, 2015). In addition, a home study is also required for non-relative spon- sors seeking to sponsor multiple children and for sponsors seeking to sponsor children who are under 12 years. Thus, the care provider uses the information collected about the child to assess each case to deter- mine whether the adult sponsors must undergo a home study prior to receiving a child in their home, in keeping with the mandates of the Trafficking Victims Protection Reauthorization Act (TVPRA) of 2008. Yet, ORR has been criticized for not adequately screening sponsors or other adults in the home prior to placement, and for not making post- placement visits, after a number of unaccompanied children were placed with human traffickers in 2015 (U.S. Senate, 2016). Once chil- dren are placed with sponsors, they are essentially lost to follow-up as ORR does not collect data on children post-placement, nor do they re- quire post-release services if declined by the sponsor (U.S. Senate, 2016). This lack of data makes it difficult to track outcomes for unac- companied children once they are released to sponsors, and the respon- sibility for collecting data on unaccompanied children falls on VOLAGs and their partner agencies in the community.
Table 2 Bivariate comparisons – experiencing a placement change.
Placement change (n = 68)
No placement change (n = 188)
M(SD) or % M(SD) or %
Age (years) 15.7 (2.4) 16.1 (2.2) Gender (1 = female) 26.47 23.81 Northern Triangle country (vs. other)⁎⁎ 69.12 76.19 Abandoned in home country 27.94 25.93 Experienced violence in home country 13.24 21.16 Failed family reunification⁎ 19.12 32.28 Fear of returning to home country 2.94 1.59 Trauma (not related to journey)⁎ 14.71 28.57 Trauma (related to journey) 7.35 8.47 Significant acting out 11.76 3.17
⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.
4. Correlates of placement stability
Much has been written in the domestic child welfare literature on the causes and consequences of foster care placement changes (ACF, 2013; Rubin et al., 2007) – but almost nothing is known about how un- accompanied children fare while in foster care. Pine and Drachman (2005) suggest that immigrant children and families experience “cumu- lative stress” related to pre-migration and transit experiences that may exacerbate the acculturation process (p. 547). Given the exposure to vi- olence and trauma experienced by unaccompanied migrant children in the U.S., maintaining placement stability in family-based settings may be difficult even with licensed, monitored and trained foster parents (Carlson et al., 2012). For example, in a study of unaccompanied Suda- nese children in the U.S., Luster, Saltarelli, Rana, Qin, and Bates (2009) found high rates of placement disruption, often linked to cultural mis- understanding between children and foster families, as well as children’s trauma histories and difficulties navigating autonomy, au- thority and trust. Children’s maltreatment histories in the general child welfare population are often linked with placement disruptions (Aarons et al., 2010), with evidence that placement changes also impact children’s functioning (Rubin et al., 2007). Not only do many unaccom- paniedmigrant children experience maltreatment and other significant forms of trauma, these children’s cultural assumptions may differ sub- stantially from those of foster families in the U.S. (Luster et al., 2009) and add a further layer of complexity to these placements in adequately meeting their needs.
Beyond the influences of trauma and culture, the impermanent legal status of unaccompanied youth also serves as a perpetual backdrop to placement stability (Pine & Drachman, 2005). As noted by the United States Conference of Catholic Bishops, “obtaining immigration relief for many unaccompaniedminors is a key goal of assistance because im- migration status affects permanency” (USCCB, 2013, p. 54). The limited legal relief options and resources available to this populationmay influ- ence children’s well-being and permanency while in care (American Immigration Council, 2015). While ORR is mandated to provide legal representation for unaccompanied children, many children are not able to receive adequate legal representation given the limited number of pro-bono attorneys available (Chen & Gill, 2015). In addition to the legal jeopardy this lack of resources places on children, the imperma- nence of legal status may have psychosocial effects. For example, a study on asylum-seeking minors from Afghanistan conducted in Swe- den showed that the unknown state of their legal status was of great concern to many of the unaccompanied children in the study (Thommessen, Corcoran, & Todd, 2015). Such unknown or imperma- nent states may also factor into the well-being of unaccompanied chil- dren and thus affect placement stability while in LTFC as they await legal adjudication.
The surge of unaccompanied children migrating to the U.S., particu- larly from Central America, has been well publicized in the media (Kristoff, 2016) and a variety of reports have examined the reasons for migration (UNHCR, 2014) and the legal and humanitarian implications for the U.S. as a receiving country (Byrne & Miller, 2012; Carlson & Gallagher, 2015). Yet, few studies (Roth & Grace, 2015) have conducted empirical examinations of how unaccompanied migrant children fare once in the U.S., and none has examined outcomes for these children while in the ORR foster care system. The current study examines place- ment stability for unaccompanied children in long term foster care from 2012 to 2015, and is guided by the following research questions: (1) To what extent do pre-migration and transit experiences influence place- ment stability? (2) To what extent do post-placement experiences and behaviors influence placement stability? and (3) To what extent are these factors similar or different in children from Northern Triangle countries (El Salvador, Guatemala, andHonduras) compared to children migrating from other countries?
The sampling frame for this study included all children served in long term foster care in the U.S. from 2012 to 2015 under the auspices of Lutheran Immigration and Refugee Service (LIRS). LIRS is an ORR-
96 T.M. Crea et al. / Children and Youth Services Review 73 (2017) 93–99
contracted implementing partner and one of themain providers of long term foster care for unaccompanied children in the U.S. During the time frame covered by this study, LIRS contracted with 5 service providers around the U.S. to provide foster homes for unaccompanied children awaiting Immigration Court hearings. Each quarter, these agencies sent standardized spreadsheets to LIRS containing a variety of informa- tion on each child in their care. For the purposes of this study, re- searchers collected all quarterly spreadsheets from LIRS and created a concatenated database for analysis. Researchers also obtained another database from LIRS related to demographic and pre-migration informa- tion on children served. These databases were linked using a common identifier, for a sample of n = 257 children. One case was missing data on all risk factors and was removed, for an analytic sample of n = 256. All procedures were approved by the university IRB.
These children were placed in the following states: Massachusetts (43.2%); Michigan (37.3%); Oregon (14.4%); and Pennsylvania (5.1%). Of the 256 children in the analytic sample, 190 (74.2%) were from the Northern Triangle countries of El Salvador, Guatemala, and Honduras, while 66 (25.8%) were from other countries including Mexico (n = 31), Somalia (n= 6), India (n=5), Nigeria (n= 5), Ghana (n=3), Ba- hamas (n= 2), Bangladesh (n= 2), Ecuador (n= 2), Ethiopia (n= 2), Haiti (n = 2), and n = 1 each for Belize, China, Mali, Nicaragua, Russia, and Saudi Arabia. As shown in Table 1, children from the Northern Tri- angle were significantly older on average (M = 16.2, SD = 1.6) than children from other countries (M = 15.2, SD = 3.6; p b 0.05). Signifi- cantly more children from the Northern Triangle experienced a failed family reunification (34.0%) compared with children from other coun- tries (13.7%; p b 0.01). Fewer children from Northern Triangle countries feared returning to their home countries (15.7%) compared with those from other countries (28.8%; p b 0.05). No other relationships were sta- tistically significant between Northern Triangle and other countries.
The dependent variable in this study is placement change, i.e., whether the child experienced a change of foster care placements while in care (vs. no placement change). Demographic variables include age at placement (in years), gender (female = 1), and country of origin (recoded to Northern Triangle country vs. other). LIRS and partner agencies collect 48 indicators on children’s behavior and experiences prior to and during placements, on topics related to childmaltreatment, behavioral issues, disabilities, criminal activities, family reunification, and medical and mental health issues. LIRS provides training and tech- nical assistance to implementing partners, as well as a guide that pro- vides definitions related to each risk factor.
Pre-migration risk factors are collected by case managers, clinicians, and program managers during intake, based on minors’ self-reports as well as referral materials. Each of these is measured as a dichotomous
Table 3 Predictors of placement change.
B(SE) OR CI
Intercept −1.44 (0.74) 0.24 – Age (years) −0.05 (0.05) 0.95 0.87–1.04 Gender (1 = female)⁎ 0.49 (0.24) 1.63 1.02–2.60 Northern Triangle country (vs. other) 0.18 (0.26) 1.20 0.72–1.98 Abandoned in home country 0.36 (0.25) 1.43 0.88–2.33 Experienced violence in home country⁎⁎⁎ 0.84 (0.24) 2.31 1.44–3.29 Fear of returning to home country⁎⁎ −0.92 (0.34) 0.40 0.20–0.78 Trauma (not related to journey)⁎⁎ −1.07 (0.33) 0.34 0.18–0.66 Trauma (related to journey) −0.24 (0.50) 0.79 0.30–2.09 Significant acting out⁎⁎⁎ 1.30 (0.34) 3.66 1.87–7.17
⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.
measure (yes = 1, no = 0) with specific definitions provided for each indicator to guide record-keeping. These indicators are updated every 90 days, as it is anticipated that initial intake interviews may provide fewer details about the minor’s journey and country of origin experi- ences. Therefore, further interactions between minors and agency staff may uncover further information that will then be recorded in the data- base. Post-migration risk factors are also collected by casemanagers, cli- nicians, and programmanagers every 90 days, based on the experiences of the minors while in care.
Independent measures in this study are the following indicators: abandoned in home country (defined as “Parent’s/guardian’s identity or whereabouts are unknown, the child has been left by the parent/ guardian in circumstances in which the child suffers serious harm, or the parent/guardian has failed to maintain contact with the child or to provide reasonable support for a significant period of time”); experi- enced violence in home country (defined as “The exposure to acts of in- terpersonal violence outside the family”); failed family reunification (with parents, relatives, or friends of the family) (defined as “Reunifica- tion with a potential sponsor began but ORR denied release; OR minor has moved from ORR designated sponsor to other caregiver”); fear of returning to home country (defined as “Minor demonstrates or ex- presses a sense of fear of returning to the home country. Or the minor fled home country due to fear of loss of life”); experienced trauma (not related to the migration journey) (defined as “Event, or series of events, or set of circumstances that is experienced as physically or emo- tionally harmful or threatening and that has lasting adverse effects on the individual’s functioning and physical, social, emotional, or spiritual well-being”); experienced trauma (related to the migration journey) (defined as “The subjection of a child, while on the journey to the US, to any act that constitutes non-accidental physical injury, emotional, verbal, physical and sexual abuse. Trauma includes both physical and psychological impact to the child. Psychological trauma includes kid- napping, hostage situations and exposure to violence, threats, death of others, etc.”); and significant acting out while in foster care (defined as “Serious behavioral problems while in foster care, including physical aggression, sexual acting out, stealing, verbal threats, running away, or other behaviors that threaten the safety of self or others”).
The analytic approach used bivariate and multivariate analyses. In- dependent samples t-tests and chi-square analyses were used to exam- ine differences in independent variables by Northern Triangle countries of origin (vs. other countries of origin), as well as by experiencing a placement change (vs. no change). Binomial logistic regression was employed to examine the relationship of independent variables to the log likelihood of ever changing placements while in foster care. Interac- tions were systematically tested between the Northern Triangle variable and all other covariates; these interactions revealed no statisti- cally significant relationships.
This study has limitations. Longitudinal analyses were not possible using these data, even though data were collected on a quarterly basis from 2012 to 2015, because the time at which a placement change oc- curred was not recorded in spreadsheets. The implications of this limi- tation are that the length of time a child spent in a placement is unable to be captured, and that the number of placement changes oc- curring during a particular quarter is also unable to be measured; both of these variables are important measures of placement stability that cannot be accounted for in the analyses. As with all administrative data, the data used for these analyses are of unknown reliability and va- lidity. A related limitation is that the indicators used by agencies to col- lect data were dichotomous and, although explicitly defined, variation in the interpretation of these indicators is possible across workers.
97T.M. Crea et al. / Children and Youth Services Review 73 (2017) 93–99
Several indicators of risk may be underreported by children who do not feel comfortable disclosing sensitive information to strangers. Indicators related to children’s protective factors were not available for analysis, but given the resilience shown by many of these children during their migration journey, future studies should include analysis of protective factors. These datamay not be generalizable to unaccompanied children placed in foster care outside of the LIRS network. Data were not avail- able on important alternative explanations of placement stability, such as the quality and type of foster placements, or variations in services provided to children across sites.
Of the 256 children in the analytic sample, 68 (26.5%) experienced a placement change while in foster care while 188 (73.4%) did not (see Table 2). Children’s age and gender did not differ to a statistically signif- icant degree by whether they had experienced a placement change. Fewer children from Northern Triangle countries (69.1%) experienced a placement change compared with those from other countries (76.2%; p b 0.01). Experiencing a failed family reunification (with par- ents, relatives, or family friends) was less common in those who had changed placements (19.1%) compared with not changing placements (32.3%; p b 0.05). Fewer children who had changed placements experi- enced trauma unrelated to the migration journey (14.7%) compared with thosewhohad not changed placements (28.6%). No other relation- ships were statistically significant.
Table 3 displays results of a binomial logistic regression model predicting the log likelihood of changing placements. Girls were 63.0% more likely to change placements than boys (OR = 1.63, CI = 1.02– 2.60, p b 0.05). Having experienced violence in their home countries is associated with over twice the likelihood of changing placements (OR = 2.31, CI = 1.44–3.29, p b 0.01). However, fear of returning to their home countries (OR = 0.40, CI = 0.20–0.78, p b 0.001) and experiencing trauma not related to migration (OR = 0.34, CI = 0.18– 0.66, p b 0.01) were each associated with a significant decrease in the likelihood of changing placements. Significant acting out while in care predicted a 266.0% greater likelihood of changing placements (OR = 3.66, CI = 1.87–7.17, p b 0.001).
This study explored the factors that influence foster care placement changes for unaccompanied migrant children, an important aspect of their experience in care while they await legal adjudication. These re- sults suggest that both pre-migration experiences and post-placement behaviors during the resettlement process are the strongest predictors of placement stability in foster care, in support of the “stage of migra- tion” framework (Pine & Drachman, 2005). Specifically, youth’s experi- ence of violence in their home countries, aswell as behavioral acting out while in care, are the strongest predictors of changing foster care place- ments. The experience of violence among these children supports existing literature that shows high rates of exposure to violence in Cen- tral American countries (UNHCR, 2014). The traumatic events which likely motivated migration are also continuing their effects in reducing placement stability, which in turn may negatively impact future well- being (Rubin et al., 2007). This pattern is also evident among refugee youth in host countries (Almqvist & Broberg, 1999; Jensen et al., 2014) as they struggle to acclimate to their host countries. The current study is unique, however, in that children in this sample are perhaps more vulnerable than children who have formal refugee status – these chil- dren are awaiting legal status hearings that may result in a permanent legal relief option, or alternatively, in immediate deportation (Byrne & Miller, 2012). The insecurity and ambiguity of this status may add fur- ther anxiety to children’s acclimation (Thommessen et al., 2015), a dy- namic that may also help trigger behavioral acting out and related placement changes, as seen in the current study. This finding is also
supported by existing research on the domestic child welfare program, which shows that there is a statistically significant relationship between placement stability and the degree of behavioral disturbance (James, Landsverk, & Slymen, 2004).
Contrary to our hypothesis and existing literature (Griffin et al., 2014; Pine & Drachman, 2005), experiencing transit-related trauma did not emerge as a predictor of placement change, and were only re- ported by fewer than 10% of the sample. At least two scenarios may help explain this pattern. First, it is possible that themagnitude of the vi- olence experienced in home countries overshadows any trauma experi- enced during themigration journey.More likely, however, childrenmay not have established a level of trust and comfort with detention screeners and caseworkers in the US to share a deep level of detail about their experiences ofmigration (Roth&Grace, 2015). If this second scenario is true, the levels of migration-related trauma may in fact be much higher.
Another counterintuitive finding is the relationship between experiencing non-migration trauma and a greater likelihood of place- ment stability. Bivariate analyses showed that nearly 15.0% of the place- ment change group, and nearly 29.0% of the non-placement change group, reported trauma not related to themigration journey. These pat- terns are similar, albeit in higher proportions, to experiencing violence in the home country, and a post hoc, bivariate correlation analysis re- vealed a weak but positive correlation between these two variables (r = 0.158, p b 0.05). Thus, one explanation for the relationship be- tween non-migration trauma and placement stability is a problem of measurement – that is, asking unaccompanied children to differentiate between their direct experiences of violence, and the psychological ef- fects of those experiences (i.e., trauma), may prove too difficult and may also imply a cultural bias of understanding the concept of trauma. Indeed, existing research on trauma-focused interventions with refugee populations show that trauma-informed interviews may not be clearly defined and articulated for refugees from various cultural backgrounds (d’Ardenne, Farmer, Ruaro, & Priebe, 2007) and that different types of trauma require different lines of conceptualization and questioning (Nickerson, Bryant, Silove, & Steel, 2011). Alternatively, it may be that experiencing trauma prior to the journey, but unrelated to the journey itself, somehow serves as a protective factor in terms of placement sta- bility and perhaps is a sign of resilience for children who have already taken the initiative to migrate without parents or caregivers.
Few unaccompanied children reported being afraid of returning to their home country, but this fear significantly predicted greater place- ment stability. Existing research has also shown that refugees fear returning to home countries related to fears for their personal safety (Goodman, Burke, Leibling, & Zasada, 2014). This dynamic likely applies to this sample of unaccompanied children, as post-hoc analyses showed that fear of returning to home countrieswasmoderately but significant- ly correlated with experiencing violence in home country (r = 0.258, p b 0.001). This finding also supports existing literature which indicates that violence caused by gangs, drug cartels, and government actors is a primary push factor for unaccompanied children migrating from the Northern Triangle and Mexico (American Immigration Council, 2015; UNHCR, 2014). Given thefindings of themultivariate analysis, fear of re- turn may serve as a motivating factor to avoid trouble while in foster care, and further legal jeopardy. In support of this argument, additional post-hoc analysis showed a negative but non-statistically significant re- lationship between fear of return and significant acting out.
In this sample, girls are more at risk of placement change than boys, although this pattern was not statistically significant in the bivariate analysis. Post-hoc analyses revealed no statistically significant relation- ships between gender and risk factors. Research has shown that girls tend to internalize emotions more (Chaplin & Aldao, 2013), and these behaviors may emerge as self-harm or suicidal ideation or behaviors which would necessitate a placement change to a higher level of care. If this hypothesis is true, girls’ behaviors are emerging in way that would not be captured by the category of “significant acting out”, as
98 T.M. Crea et al. / Children and Youth Services Review 73 (2017) 93–99
this variable implies externalizing behaviors.More research is needed to identify whether unaccompanied girls are at higher risk of placement instability.
Some notable differences emerged between children from the Northern Triangle and other countries. Significantly more children from non-Northern Triangle countries feared returning home, even though over 1/4th of children experienced violence in their home coun- tries regardless of their origin. This difference may simply reflect differ- ing contexts, such as poverty (Mexico), war (Somalia), or gang-related violence (Northern Triangle countries), or may also reflect cultural dif- ferences in what information children feel comfortable reporting, and to whom. Significantly more Northern Triangle children experienced a failed family reunification prior to entering foster care. This finding re- flects a common driver of migration for these children to reunite with their birth families and relatives (Byrne & Miller, 2012; Chishti & Hipsman, 2015), a dynamic which is likely much less pronounced in children from the wide range of other countries represented in this sample.While children fromNorthern Triangle countrieswere less like- ly to experience a placement change in bivariate analyses, this relation- ship disappeared in the multivariate analyses when holding other variables constant. Other variables, such as behavioral problems, proved to be more significant in predicting the likelihood of a placement change, regardless of country of origin. This finding thus is consistent with research on placement changes in domestic child welfare, al- though, as with existing research, it is difficult to disentangle whether behavior precipitates the placement change, or whether the placement change precipitates the behavior (Rubin et al., 2007).
8. Conclusion and implications
As unaccompanied children migrate to the United States in consis- tent numbers (CBP, 2016), it is imperative that those in social services be well-informed and prepared to address the multiple needs of this population. In some ways, given the ORR-funded child welfare system, the challenge of serving these children provides an opportunity for the convergence of two previously separate fields of study in the US: child welfare, and refugee studies. Drawing from evidence in both fields, re- searchers and practitioners can explore how best to define childwelfare outcomes in terms of safety, permanency, and well-being, as applied to children placed in foster care because of their migration and legal sta- tuses, rather than for reasons related to maltreatment alone. This emerging dynamic also reinforces the need to strengthen ongoing data collection and reporting efforts, to provide service providers crucial information about the needs of this vulnerable population.
As the current study shows, the pre-migration and resettlement ex- periences of unaccompanied children may significantly impact their permanency outcomes in U.S. foster care. The effects of cumulative stress experienced by many unaccompanied children across the stages of migration (Pine & Drachman, 2005) will unfold in the context of U.S. social service systems. It is therefore critical that social workers and other professionals consider the multilayered experiences of these children, which are unique compared to children in domestic child welfare systems. Existing research shows that Latino children of im- migrant parents are in greater need of mental health services than Latino children of U.S.-born parents (Dettlaff & Cardoso, 2010). The current study adds evidence that unaccompanied migrant chil- dren are also likely in need of such services while placed in foster care, and that addressing behavioral disturbances may help increase placement stability.
It is also important that social service agencies operate from cultur- ally competent, trauma-informed perspectives, such that caseworkers and foster parents are well educated and prepared to meet the unique needs of these children. Dettlaff and Rycraft (2010) describe a sys- tems-of-care approach to child welfare practice specific to immigrant Latino families, which is responsive to and respectful of cultural values and differences. Central to this effort is the need for bicultural and
bilingual social workers equipped with the skills to help engage immi- grant children and families in navigating the complexities of the U.S. so- cial service landscape. A trauma-informed perspective of care can also help professionals interpret youth’s behavior in care and respond to this behavior effectively (Bath, 2008). Yet, the concepts of trauma and mental health may be culturally bound constructs, such that further work should be done to interpret or revise these constructs in light of cultural differences and values (Gozdziak, 2004).
Future research for unaccompanied children requires a clear articu- lation of outcomes related to child safety, permanency, and well-being (Dettlaff & Rycraft, 2010). This articulation is complicated by limitations in existing standardized measures. For example, the Harvard Trauma Questionnaire (HTQ; Mollica et al., 1992) is a commonly used instru- ment to identify post-traumatic stress disorder in refugee populations. This measure has also been used to examine mental health outcomes of refugees seeking treatment (Knipscheer, Sleijpen, Mooren, Heide, & van der Aa, 2015), and may be applicable to migrant children served in U.S. foster care for the purposes of trauma-informed treatment plan- ning. These data could also be useful to researchers in examining trajec- tories and outcomes for unaccompanied children as they navigate the legal system, transition out of foster care, and begin integrating in Amer- ican society. Yet, the HTQ has also been criticized as not reflecting the full breadth of migration experiences commonly faced by unaccompa- nied youth, particularly from Central America (Cardoso et al., 2016). Fu- ture research and practice could benefit by developing assessments, or validating existing assessments, that account for cultural differences, values and experiences of unaccompanied youth.
A larger issue for unaccompanied children in foster care involves the different and often conflicting mandates of child welfare and immigra- tion systems. Child welfare systems are focused on establishing safety, permanency and well-being for children in care. Yet, unaccompanied youth are normally placed in LTFC for reasons of migration rather than maltreatment, and child welfare outcomes for this population remain largely undefined. Even in the context of the current study,where place- ment stability serves as ameasure of permanency, all of the youth in the sample are characterized by lack of legal status, and therefore imperma- nency while they await legal adjudication. Further, as Dettlaff and Cardoso (2010) have noted, immigrant children involved in child welfare face a number of barriers beyond cultural and linguistic differences. Increasing globalization and patterns of transnational migration have resulted in a backlash of anti-immigrant policies in the U.S. that threaten to curtail existing supports considerably. Yet, professionals in child welfare must continue with their mandate to pursue safety, permanency, and well-being for unaccompanied and immigrant youth in care.
As rates of migration continue to increaseworldwide, and unaccom- panied children continue to migrate in consistently large numbers, the field of childwelfare could benefit fromadopting amore global perspec- tive. Barn, Križ, Poso, and Skivenes (2014) coined the term “global child protection” to describe this dynamic, in which issues of child protection can no longer be considered as confined within the boundaries of na- tions but are based on the universal rights of children. Global child pro- tection is focused on the principles that a child’s legal status does not affect their rights for protection; that migrant children who fled harm or risk of harm in their countries of origin still need protection even if after leaving the situation that placed that at risk; and that those work- ing in child protection policy and practice need to develop transnational cultural competency, an understanding of discrimination against mi- grant children, and an understanding of children’s vulnerabilities relat- ed to pre-migration andmigration experiences. As the current study has shown, youth’s experiences in their countries of origin are implicated in their functioningwhile in their host country. It is therefore necessary for child protection professionals and researchers to consider the condi- tions experienced bymigrant children prior to arrival, as well children’s underlying cultural values, because these conditions and values frame children’s current functioning.
99T.M. Crea et al. / Children and Youth Services Review 73 (2017) 93–99
This studywas funded by Lutheran Immigration and Refugee Service (LIRS) (Project #5102321).
Aarons, G. A., James, S., Monn, A. R., Raghavan, R., Wells, R. S., & Leslie, L. K. (2010). Behav- ior problems and placement change in a national child welfare sample: A prospective study. Journal of the American Academy of Child and Adolescent Psychiatry, 49(1), 70–80.
Administration for Children and Families (2012). Office of refugee resettlement: Voluntary agencies. Available from: http://www.acf.hhs.gov/programs/orr/resource/voluntary- agencies.
Administration for Children and Families (2013). Child welfare outcomes 2010–2013: Re- port to congress. Available from: https://www.acf.hhs.gov/sites/default/files/cb/ (cwo10_13.pdf#page = 25).
Almqvist, K., & Broberg, A. (1999). Mental health and social adjustment in young refugee children 3½ years after their arrival in Sweden. Journal of the American Academy of Child and Adolescent Psychiatry, 38(6), 723–730.
American Immigration Council (2015). A guide to children arriving at the border: Laws, policies and responses. Available from https://www.americanimmigrationcouncil. org/research/guide-children-arriving-border-laws-policies-and-responses.
Barn, R., Križ, K., Poso, T., & Skivenes, M. (2014). Migrant children and child welfare sys- tems: A contested challenge. In M. Skivenes, R. Barn, K. Kriz, & T. Pösö (Eds.), Child welfare systems and migrant children: A cross country study of policies and practice. Ox- ford, UK: Oxford University Press.
Barth, R. P. (2002). Institutions vs. foster homes: The empirical base for the second century of debate. Chapel Hill, NC: UNC School of Social Work, Jordan Institute for Families.
Barth, R. P., & Jonson-Reid, M. (2000). Outcomes after child welfare services: Implication for the design of performance measures. Children & Youth Services, 22(9–10), 763–787.
Bath, H. (2008). The three pillars of trauma-informed care. Reclaiming Children and Youth, 17(3), 17–21.
Bean, T., Eurelings-Bontekoe, E., & Spinhoven, P. (2007). Course and predictors of mental health of unaccompanied refugee minors in the Netherlands: One year follow-up. Social Science & Medicine, 64(6), 1204–1215.
Byrne, O., & Miller, E. (2012). The flow of unaccompanied children through the immigration system: A resource for practitioners, policy makers, and researchers. Center on Immigra- tion and Justice, Vera Institute of Justice Available from: http://archive.vera.org/sites/ default/files/resources/downloads/the-flow-of-unaccompanied-children-through- the-immigration-system.pdf.
Cardoso, J. B., Brabeck, K., Stinchcomb, D., Heidbrink, L., Price, O. A., Gil-Garcia, O. F., … Zayas, L. H. (2016). Challenges to integration for unaccompanied migrant youth in the post-release U.S. context: A call for research. (Unpublished manuscript available from authors).
Carlson, E., & Gallagher, A. M. (2015). Humanitarian protection for children fleeing gang- based violence in the Americas. Journal onMigration and Human Security, 3(2), 129–158.
Carlson, B., Cacciatore, J., & Klimek, B. (2012). Risk and resilience perspective on unaccom- panied refugee minors. Social Work, 57(2), 259–269.
Chaplin, T. M., & Aldao, A. (2013). Gender differences in emotion expression in children: A meta-analytic review. Psychological Bulletin, 139(4), 735–765.
Chen, A., & Gill, J. (2015). Unaccompanied children and U.S. immigration system: Chal- lenges and reforms. Journal of International Affairs, 68(2), 115–133.
Chishti,M., & Hipsman, F. (2015). The child and familymigration surge of summer 2014: A short-lived crisis with a lasting impact. Journal of International Affairs, 68(2), 95–114.
Crea, T. M., Barth, R. P., & Chintapalli, L. (2007). Home study methods for evaluating pro- spective resource families: History, current challenges, and promising approaches. Child Welfare, 86(2), 141–159.
d’Ardenne, P., Farmer, E., Ruaro, L., & Priebe, S. (2007). Not lost in translation: Protocols for interpreting trauma-focused CBT. Behavioural and Cognitive Psychotherapy, 35, 303–316.
Derluyn, I., Mels, C., & Broekaert, E. (2009). Mental health problems in separated refugee adolescents. Journal of Adolescent Health, 44(3), 291–297.
Dettlaff, A. J., & Cardoso, J. B. (2010). Mental health needs and service use among Latino children of immigrants in the child welfare system. Children and Youth Services Review, 32, 1373–1379.
Dettlaff, A. J., & Rycraft, J. R. (2010). Adapting systems of care for childwelfare practice with immigrant Latino children and families. Evaluation and Program Planning, 33, 303–310.
Duerr, A., Posner, S. F., & Gilbert, M. (2003). Evidence in support of foster care during acute refugee crises. American Journal of Public Health, 93(11), 1904–1909.
Goodman, S., Burke, S., Leibling, H., & Zasada, D. (2014). ‘I can’t go back because if I go back I would die’: How asylum seekers manage talk about returning home by highlighting the importance of safety. Journal of Community and Applied Social Psychology, 25(4), 327–339.
Gozdziak, E. M. (2004). Training refugee mental health providers: Ethnography as a bridge to multicultural practice. Human Organization, 63(2), 203–210.
Griffin, M., Son, M., & Shapleigh, E. (2014). Children’s lives on the border. Pediatrics, 133, e1118.
James, S., Landsverk, J. A., & Slymen, D. J. (2004). Patterns of placement movement in out- of-home care: Patterns and predictors. Children and Youth Services Review, 26(2), 185–206.
Jensen, T., Skårdalsmo, E., & Fjermestad, K. (2014). Development of mental health prob- lems – A follow-up study of unaccompanied refugee minors. Child and Adolescent Psychiatry and Mental Health, 8(29). http://dx.doi.org/10.1186/1753-2000-8-29.
Keles, S., Friborg, O., Idsøe, T., Sirin, S., & Oppedal, B. (2015). Depression among unaccom- paniedminor refugees: The relative contribution of general and acculturation-specific daily hassles. Ethnicity & Health, 21(3), 1–18.
Kirmayer, L., Narasiah, L., Munoz, M., Rashid, M., Ryder, A., Guzder, J., & Pottie, K. (2011). Common mental health problems in immigrants and refugees: General approach in primary care: Canadian Guidelines for Immigrant Health. CMAJ: Canadian Medical Association Journal, 183(12), E959.
Knipscheer, J. W., Sleijpen, M., Mooren, T., Heide, F. J. J., & van der Aa, N. (2015). Trauma exposure and refugee status as predictors of mental health outcomes in treatment- seeking refugees. BJPsych Bulletin, 39, 178–182.
Kristoff, N. (2016). We’re helping to deport kids to die. New York Times, July 16, 2016. Available from: http://www.nytimes.com/2016/07/17/opinion/sunday/were-helping- deport-kids-to-die.html?_r=0.
Luster, T., Saltarelli, A., Rana, M., Qin, D., & Bates, L. (2009). The experience of Sudanese unaccompanied minors in foster care. Journal of Family Psychology, 23(3), 386–395.
Mollica, R. F., Caspi-Yavin, Y. P., Bollini, P., Truong, T., Tor, S., & Lavelle, J. (1992). Validating a cross-cultural instrument for measuring torture, trauma, and post-traumatic stress- disorder in Indochinese refugees. The Journal of Nervous and Mental Disease, 180(2), 111–116.
Newton, R. R., Litrownik, A. J., & Landsverk, J. A. (2000). Children and youth in foster care: Disentangling the relationship between problem behaviors and number of place- ments. Child Abuse and Neglect, 24(10), 1363–1374.
Nickerson, A., Bryant, R. A., Silove, D., & Steel, Z. (2011). A critical review of psychological treatments of posttraumatic stress disorder in refugees. Clinical Psychology Review, 31, 399–417.
Office of Refugee Resettlement (2015). Children entering the United States unaccompanied. Available from: http://www.acf.hhs.gov/programs/orr/resource/children-entering-the- united-states-unaccompanied.
Office of Refugee Resettlement (2016a). Children entering the United States unaccompa- nied: Guide to terms. Available from: http://www.acf.hhs.gov/orr/resource/children- entering-the-united-states-unaccompanied-guide-to-terms.
Office of Refugee Resettlement (2016b). Facts and data. http://www.acf.hhs.gov/orr/ about/ucs/facts-and-data.
Office of Refugee Resettlement (2016c). Unaccompanied children released to sponsors by state. http://www.acf.hhs.gov/orr/programs/ucs/state-by-state-uc-placed-sponsors.
Pine, B. A., & Drachman, D. (2005). Effective child welfare practice with immigrant and refugee children and their families. Child Welfare, 84(5), 537–562.
Roth, B. J., & Grace, B. L. (2015). Falling through the cracks: The paradox of post-release services for unaccompanied child migrants. Children and Youth Services Review, 58, 244–252.
Rubin, D. M., O’Reilly, A. L. R., Luan, X., & Localio, A. R. (2007). The impact of placement stability on behavioral well-being for children in foster care. Pediatrics, 119(2), 336–344.
Sourander, A. (1998). Behavior problems and traumatic events of unaccompanied refugee minors. Child Abuse & Neglect, 22(7), 719–727.
Thommessen, S. A. O., Corcoran, P., & Todd, B. K. (2015). Experiences of arriving to Swe- den as an unaccompanied asylum-seekingminor from Afghanistan: An interpretative phenomenological analysis. Psychology of Violence, 5(4), 374–383.
U.S. Customs and Border Protection (2016). United States Border Patrol Southwest family unit subject and unaccompanied alien children apprehensions, Fiscal Year 2016. Avail- able from: https://www.cbp.gov/newsroom/stats/southwest-border-unaccompanied- children/fy-2016.
U.S. Senate (2016). Protecting unaccompanied alien children from trafficking and other abuses: The role of the Office of Refugee Resettlement. United States Sen- ate, Permanent Subcommittee on Investigations, Committee on Homeland Secu- rity and Governmental Affairs. Available from: http://www.hsgac.senate.gov/ subcommittees/investigations/hearings/adequacy-of-the-department-of-health-and- human-services-efforts-to-protect-unaccompanied-alien-children-from-human- trafficking.
United States Conference of Catholic Bishops (USCCB) (2013). TheUnited States Unaccom- panied RefugeeMinor program: Guiding principles and promising practices. Available from http://www.usccb.org/about/children-and-migration/unaccompanied-refugee- minor-program/upload/united-states-unaccompanied-refugee-minor-program- guiding-principles-and-promising-practices.pdf.
United Nations High Commissioner for Refugees (UNHCR) (2014). Children on the run: Unaccompanied children leaving Central America and Mexico and the need for inter- national protection. Available from http://www.unhcrwashington.org/sites/default/ files/1_UAC_Children%20on%20the%20Run _Full%20Report.pdf.
Women’s Refugee Commission (2012). Forced from home: The lost boys and girls of Central America. Available from: https://www.womensrefugeecommission.org/uncategorized/ 2057-forced-from-home-the-lost-boys-and-girls-of-central-america-background-and- report.