The EMDR Integrative Group Treatment Protocol in a Psychosocial Program for Refugee Children: A Qualitative Pilot Study

Research Design and Ethics

The qualitative case study design involved children completing outcome and session rating scales at the beginning and end of each session and a SUD scale for traumatic memories before and after the EMDR-IGTP session. Therapists reflected on program process and outcomes through completing a postprogram report. To better understand emergent cultural challenges, a focus group of four Arab interpreters was held. Ethical approval was granted by the CHUMS (Community Interest Company) ethics review board and by the University Research Ethics Committee involving active signed informed assent/consent from children and families, therapists, and interpreters.

Organizational and Geographical Context

The current project was a collaboration between two social enterprise companies: (a) Bevan Healthcare, providing medical assessments and treatment for families in need, and (b) CHUMS, who provide a range of child mental health services in the East of England, including emotional well-being, trauma, and traumatic bereavement services. CHUMS comprise a multiprofessional team experienced in working with trauma and empirically based interventions. The project took place in West Yorkshire, a region with an ethnically diverse population of 531,200, ranked the 19th most economically deprived local authority in England. Six hundred and fifty refugees were received into the area in the last 3 years from Afghanistan, Iraq, Iran, Syria, Eritrea, the Democratic Republic of Congo, Sudan, and Ethiopia. Child refugees had suffered significant trauma from war, bereavement, loss of friends, family and others left behind, the difficulty of coming to a new culture, isolation, language and cultural differences, racism (in their home countries and the United Kingdom), trafficking, rape and sexual abuse, gender-based violence, torture, and imprisonment. The project explored what could be delivered to children who were unable to access statutory child mental health services that appeared ill-equipped to support children with such traumas and language and cultural barriers.

Participants

Participants were eight child refugees, two therapists, and four additional interpreters.

Psychosocial Program

CHUMS has a history of running successful group programs for children, using creative approaches to foster social support, normalization of symptoms, and meaning-making for those who have experienced trauma of bereavement (Siddaway, Wood, Schulz, & Trickey, 2015). The focus of this pilot group was to promote similar approaches to enhance the emotional well-being of refugee children exposed to trauma. The primary goals were to help children feel safe, grow in emotional awareness and resources, and process any traumatic experiences. A four-session program was developed, which incorporated EMDR-IGTP, within the second session. The sessions were of 3 hours duration and utilized art- and music-based interventions to engage children (Stallard, 2005). The program aimed to assimilate in a child-friendly way new positive associations, help the processing of traumatic memories, promote resilience for the future, and increase the sense of perceived mastery of potentially distressing emotions that were previously avoided (Solomons & Shapiro, 2008).

The combination of activities including fitness, arts, and crafts has been shown to decrease insecurity (Panter-Brick et al., 2018), and although a small effect, the authors argue such an approach is useful, particularly if other services are not available due to high thresholds for Child and Adolescent Mental Health Service teams and the cultural stigma around mental health difficulties. Music therapy is a growing area of research for anxiety, depression, and PTSD. Music-based intervention has been shown to sustain engagement over talking forms of CBT (Trimmer, 2017). Carr et al. (2012) identified the effectiveness of treating PTSD for patients who had not responded to CBT therapy, noting the primary goals were to foster safety and encourage the use of music to communicate emotional material and alleviate re-experiencing. The secondary goals of music therapy were identified as socialization, reduced anxiety, and relaxation. Research with children in Northern Ireland who presented with behavioral and emotional problems showed music therapy significantly improved their self-esteem, communication skills, and long-term social functioning (Porter et al., 2017).

Program Staff

The psychosocial group was facilitated by the aforementioned two therapists, two volunteers, and two interpreters. One volunteer was male and a general practitioner, and the other a female nurse. Both ran routine medical checks for the families at Bevan Healthcare. The two interpreters were Syrian Arabic translators, one male and one female. Both were known to the families from their work interpreting for housing and medical needs in the past. Neither had direct experience of the war in Syria and had been living in the United Kingdom for a significant period of time. Prior to program delivery interpreters completed a 3-hour training session on the signs and symptoms of trauma, a basic understanding of the problems children may have faced as well as program theoretical underpinnings, activities, and technical language. It was during this training that the structure of the day was amended to include a culturally sensitivity time for prayer.

Program Composition

Session 1 involved creating a positive, safe environment for the stabilization phase, including video clips for psychoeducation and music/cartoons as representations of emotional states encouraging the children to relate to the concept of emotions and the expressions of these through sound. Session 1 also introduced the use of BLS as drums and kalimbas when discussing and identifying emotions. Children also experienced the butterfly hug technique and belly breathing, a relaxation technique. In addition, the children were introduced to a chocolate manufacturing metaphor for good and bad memories. Good memories are represented as having the right ingredients to make them edible, bad memories therefore lack some of the ingredients and can produce unpleasant tasting chocolate or even cause a disruption to the manufacturing business. Further resources were developed using the Tree of Life (TOL; Ncube-Millo & Denborough, 2006) to reconnect the children to their country of origin and their own and family strengths as well as providing a reference for a theme of trees and bird migration which ran throughout the 4-week program. The TOL has been used in a variety of applications with children including developing cultural strengths and increasing self-concept (German, 2013).

The second session involved discovering degrees of emotion through sound volume and frequency of rhythm and drawing a safe place installed by butterfly hugs. Despite previous research (Hamdan-Mansour, Abdel Razeq, AbdulHaq, Arabiat, & Khalil, 2017), all children were able to generate a positive, safe place, including parks, playgrounds, and a mosque. The concept of bad memories and how they can make you feel was introduced. Bad memories were described using the chocolate metaphor where the factory uses the wrong ingredients, causing the machinery to break down and the need for an engineer to help fix it. This session included the delivery of EMDR-IGTP, where children were encouraged to draw their traumatic memories and reprocess these through tapping. Because of limited available space, all children sat around one large table. Children, with a support worker beside them, completed their drawings and used the butterfly hug individually at their own pace, until positively associated drawings were produced. The session ended with a “migratory birds” healing narrative that provided reasoned choices for leaving home to find safety elsewhere (Kurtmazler & Kurt, 2016).

Session 3 involved further emotional expression but as a group playing instruments together as a band. The children were given the opportunity to add any other resources or positive memories to their TOL and were introduced to an additional healing narrative: “the seed.” This promoted the idea of posttraumatic growth. The children drew new trees grown from seeds they were given and used these to make a shared forest to create a safe home for a cuddly “refugee” toy.

The final session was developed as a reward for attendance to ensure data collection of the outcome measure. A group “family” trip to the Yorkshire Dales was arranged. This completed the program theme introducing the families to an English forest and learning how it sustains life there. Activities included leaf collecting, memory sticks, and psychoeducation.

Measures

Arab Interpreter Focus Group

In order to understand the cultural challenges within program delivery, the principal researcher (RH) conducted a focus group of interpreters (n = 4). Interpreters were asked (a) their perception of what cultural challenges exist in the delivery of a Western program to children from an Arab culture, (b) why these challenges arise, and (c) what are the most useful ways to prepare or respond to such challenges. Notes of the focus group were recorded verbatim by the principal researcher.

Analysis

Differences in SUD scores were compared pre- and post-IGTP. RH conducted the initial thematic analysis on the therapists’ reflections contained within the project report and on focus group data. Themes were extracted using Braun and Clarke’s (2006) systematic approach to thematic analysis. That is, statements quoted in the report by therapists were listed under common themes of meaning. Using the same process, statements made by interpreters in the focus group and analyzed as a group response were listed for common themes of meaning. These themes were then checked against the narrative of the project report to ensure the names for themes provided an accurate representation of the report contents. The latter was conducted by RH and the family care worker together. A comparative analysis was then applied across therapist, interpreters’ statements, and themes to identify similarities and differences between themes across the two participant groups. This comparative analysis was initially by RH and then by an independent researcher (IB), the latter a professor in psycho-trauma studies. Inter-rater reliability of report and focus group statements and themes was provided by an independent analysis of matched themes from statements by RH and IB and is reported as a percentage of agreement of theme names. Inter-rater reliability involved checking themes against participant statements recorded under each theme within the project report.

EMDR-IGTP Treatment Responses and SUD Scores

Children attended all the sessions except for the second week where two girls from the same family did not attend because of bereavement in the community. As a consequence, the girls received EMDR-IGTP when the other children experienced an extended music activity in another room in Session 3. Prior to EMDR-IGTP, where children rated their disturbance numerically, all rated 10. Child refugees were, therefore, reporting the highest levels of internal disturbance. Pictures similarly indicated high levels of disturbance. Following EMDR-IGTP, SUD scores substantially reduced for seven of the eight children, whether communicated by numbers and/or in pictures (see Table 1). Where numbers were reported, all SUD scores reduced to zero. Post-IGTP pictures, in contrast to pre-IGTP pictures (sad faces), included smiling happy faces, again indicating a significant shift in emotion. One child spontaneously rated happiness a 10 following EMDR-IGTP.

Figures 1–4 show an example of the sequence of processing in pictures. The girl LL, aged 10 years 3 months, drew her target memory as two sad people and one lying on a bed. A sad face hovers above both of them, indicating the sadness. She rated this 10. After the first set of BLS, she drew another picture of the two people, this time happy, scoring a 5. Her third picture was more negative. Again, there were two figures: One, possibly male, was looking sad. The other figure was lying down with the head separated from the body. The therapist felt that this child required further processing to enable more positive information to be accessed rather than being left with a potentially negative memory. The girl was unable to think of another picture after BLS, so the clinician offered a cognitive interweave. He asked “How big were you when the memory had happened?” After further BLS, the girl drew another picture that showed herself when younger and sad and on the opposite side of the page a bigger version with a smiley face. After further BLS, the girl drew herself standing large next to her new school saying she was now safe in school. Again from an AIP model, this series of pictures shows how processing can become stuck with linked negative memories. Introducing the concept of time enabled her to put the events into her linear narrative and separate them from her current safety in her new school, something she enjoys considerably and describes as “happy and safe.”

Figure 1.

LL, aged 10 years 3 months, drew her target memory. SUD score = 10.

Figure 2.

The same people from LL’s target memory, but happy. SUD score = 5.

Figure 3.

LL’s third drawing.

Figure 4.

LL’s drawing showing herself on the left as younger and sad, and on the right as older and with a smiley face.

To exemplify the nature of change, more details are provided for specific children. One boy AL, aged 11 years 9 months, drew his first target memory of a hospital being bombed in Syria. His drawing showed the aircraft bombs and explosion with people shouting for help. He rated this picture a 10 on the SUD scale. After BLS, he drew his family arriving in Lebanon by plane, a positive picture, with a low SUD score of 1 out of 10. After further BLS, he drew his final picture, his dream of the future. In this picture, he drew himself as the prime minister and his family members all with professional jobs. He wrote next to this a “lovely dream” and scored his distress as zero. From an accelerated information processing model, this series of pictures indicates a movement in thought processing with new adaptive information of safety being incorporated into the bad memory facilitating a positive view of the future. Table 1 shows each participant, his or her age, and his or her reported SUD scores pre- and post-IGTP. In addition, therapist comments made following the EMDR-IGTP session and recorded in the project report are noted for each child.

Other children showed simpler processing starting with sad faces and experiences and then ending up drawing happy faces. One boy O, aged 8 years 10 months who presented throughout with the most severe symptoms, was invited to repeat EMDR-IGTP on the third week. In this session, he arrived in an agitated state, speaking quickly, and hiding behind the sofa. He was taken from the group and provided with individual EMDR-IGTP to encourage his processing during this session. He drew a series of angry and sad faces with indistinct objects. One picture appeared to be a plane dropping bombs injuring someone. This was virtually identical to the previous week’s picture. The processing continued, and his final drawing was a World. During this session, balloons were used to draw maps of Syria and the United Kingdom and how close they were. With this final drawing, this boy said “I’m happy, love the World.”

Themes in the Therapist Reflections

A high level of inter-rater reliability was achieved for the number of statements listed under the seven identified themes (58 statements vs. 59 statements equaling 98% agreement) identified from therapist reflection. The second rater IB perceived separate statements were made for the need to test the assumption for a more narrow group age range.

Themes From the Interpreter Focus Group

A high level of inter-rater reliability was achieved for the statements listed under the identified themes for the interpreters’ focus group (32 vs . 34 statements equaling 94% agreement). Four main themes were identified when reviewing the assessment and activities of the pilot group: understanding mental health, psychological and cultural language/meanings, the interpreter's role, and the importance of relationship and culture. The second rater also identified the need for a more extended discussion to be held.

Comparison Across Children, Therapists, and Interpreters

Children’s reduced SUD scores and shift from sad to happy faces for EMDR-IGTP were supported by therapists’ reports of reduced child distress. Therapists’ comments on children’s growth in emotional awareness also complemented children’s SUD scores. Therapists’ and Arab interpreters’ comments also complemented and contrasted with each other (Table 3). Both therapists and interpreters identified the benefits of the program as culturally sensitive and a nonpathological way of working. In contrast, interpreters emphasized the stigma of mental health, the importance of culturally sensitive language, the differences in communicative control, and the quality of relationships as inhibiting program effectiveness. Therapists, on the other hand, focused on interpreters as a resource, the need for good training for interpreters, and the use of music as one way of breaking down cultural barriers. Therapists also focused on client selection, engagement, assessment, and monitoring involving family and teachers.

Cultural Challenges

Both therapists and interpreters identified a wide range of cultural challenges for the whole program, rather than specifically EMDR-IGTP. These included (a) the differing conceptualizations of mental health between East and West and the problem of stigma of mental illness in Arab cultures, (b) some psychological terms did not have a direct translation into Arabic and/or there were new psychological terms and concepts that needed further exploration, and (c) a confusion of roles between therapist and interpreter, leaving interpreters wanting more information about the nature of traumatization and resultant symptomology. Therapists, on the other hand, were cautious about translators going beyond interpretation and therapists losing control of the conversation. The latter was perceived to create a possible risk of harm for traumatized children. As a useful resource, the National Child Traumatic Stress Network (NCTSN)_provides a framework to enable therapists to anticipate and plan for these and other cross-cultural issues (National Child Traumatic Stress Network, 2006). Interpreters reported that child disclosure of family information is interpreted as an act of betrayal from an Arab family perspective. Family bonds were described as closer than in the West, where family loyalty in the Middle East was expected because of fears about family reputation in the community. This was perceived to be a significant factor in children’s reluctance to share. In addition, refugees had been targeted for racial abuse in the local community, and the sense of safety for the family may have been compromised with the parents communicating their sense of insecurity, intentionally or otherwise. The group did provide an initial brief talk for parents on the goals of the group, but future programs may be better served by completing full assessments with families that include the verbal delivery of outcome measures within the context of a conversation and discussion with the aid of interpreters.

Interpreters also highlighted that children in the Middle East expect more direction with explicit step-by-step instruction, rather than the more choice-oriented way of communicating in the West. In this regard, the EMDR-IGTP sessions with a protocol structure may have been more attuned than the more free-flowing creative activities in other sessions. Finally, given the Arab focus on spoken rather than written language, routine outcome measures and screening questionnaires were seen by the Arab interpreters as a less effective way of enabling children to respond. The current study then not only affirms the value of interpreters in working with refugees but also identified a range of challenges that need to be addressed in preparation (d’Ardenne, Ruaro, Cestari, Fakhoury, & Priebe, 2007).

Limitations

Because of the novel application of EMDR-IGTP within a new psychosocial program, a qualitative design was utilized. Such a design, however, does not provide a measure of the size of change nor does it infer generalizability. Sample sizes of children, practitioners, and interpreters were small, and therefore, not all issues are likely to have been identified. Feedback from individual interpreters who had worked with the program, rather than others in a focus group, may have yielded more information on the experiences during the program. Lack of digitally recorded data raises questions of bias and reliability. While routine outcome measures were built into the program, the lack of training for interpreters meant the prescores were invalid, and the postprogram education session did not include the therapists and was run by the volunteers who were unable to complete the data collection. In the future, therapist involvement in the collection of pre- and post-questionnaires with conversations with the family is paramount for future evaluation of projects such as this. With these limitations, extendibility of research is more appropriate than generalizability of findings.

EMDR-IGTP utilizes the eight phases of EMDR with group and art therapy models. The butterfly hug is used for BLS (Jarero & Artigas, 2014; Shapiro, 2001). Studies have used a range of terms to describe EMDR-IGTP such as the EMDR group protocol, the children’s EMDR group protocol, and the butterfly hug protocol (Jarero & Artigas, 2015), which has led to some confusion of what was actually evaluated. In addition, there have been a number of adaptations to the protocol such as the IMMA protocol, which uses a bracelet as an anchor for positive resources (Laub & Bar-Sade, 2009). It appears, however, that the treatment may be very robust, with adaptations reporting good results.

Recommendations for Practice

EMDR-IGTP should continue to be embedded into and potentially expanded within an evolving psychosocial group-based program for child refugees. Further consideration needs to be given to the selection process and age range of the children. Standardized screening measures delivered prior to program delivery would facilitate the selection of children with clinical levels of PTSD, although these may be best delivered as part of an assessment conversation with parents. Groups of children with similar developmental levels may help with matching cognitive ability and program content. Interpreters need training to understand the aims, theory, and practices of EMDR-IGTP and psychosocial program delivery, including the process of change with EMDR to provide the best support for group delivery.

Recommendations for Research

Further exploration is needed of therapist, interpreter, and child experience of EMDR-IGTP. Experimental research designs would bring more rigor to analyzing the effectiveness of EMDR-IGTP as a component within psychosocial programs. Child and parental standardized outcome measures provide a measure of symptom change, for example, for posttraumatic stress, anxiety, depression, and complicated grief. Inclusion of longitudinal evaluation would assess the maintenance of gains and the development of posttraumatic growth. Given the complexity of cross-cultural groups with traumatized children, unintended negative consequences should be explicitly explored. Finally, children are more likely to suffer PTSD if their parents have also developed PTSD. These parents are less likely to notice the symptoms in their children and seek help (Meiser-Steadman, Smith, Yule, Glucksman, & Dalgeish, 2016). Further research could consider parental PTSD and how a parent component to the group may facilitate further change.