METHOD: We conducted in-depth interviews with stakeholders in Malaysia (N = 44) and Thailand (N = 50), alongside policy document review in both countries. Data were analysed thematically. Results informed development of Systems Thinking diagrams hypothesizing potential intervention points to improve cultural competency, namely via addressing language barriers.
RESULTS: Language ability was a core tenet of cultural competency as described by participants in both countries. Malay was perceived to be an easy language that migrants could learn quickly, with perceived proficiency differing by source country and length of stay in Malaysia. Language barriers were a source of frustration for both migrants and health workers, which compounded communication of complex conditions including mental health as well as obtaining informed consent from migrant patients. Health workers in Malaysia used strategies including google translate and hand gestures to communicate, while migrant patients were encouraged to bring friends to act as informal interpreters during consultations. Current health services are not migrant friendly, which deters use. Concerns around overuse of services by non-citizens among the domestic population may partly explain the lack of policy support for cultural competency in Malaysia. Service provision for migrants in Thailand was more culturally sensitive as formal interpreters, known as Migrant Health Workers (MHW), could be hired in public facilities, as well as Migrant Health Volunteers (MHV) who provide basic health education in communities.
CONCLUSION: Perceptions of overuse by migrants in a health system acts as a barrier against system or institutional level improvements for cultural competency, in an already stretched health system. At the micro-level, language interventions with migrant workers appear to be the most feasible leverage point but raises the question of who should bear responsibility for cost and provision-employers, the government, or migrants themselves.
METHODS: A systematic process was followed in formulating the therapy and devising a treatment manual consistent with the principles of the Adaptation and Development After Persecution and Trauma (ADAPT) model. The process of development and refinement was based on qualitative research amongst 70 refugees (ten from West Papua and 60 Rohingya from Myanmar). The therapeutic process was then piloted by trained interventionists amongst a purposively selected sample of 20 Rohingya refugees in Malaysia.
RESULTS: The final formulation of IAT represented an integration of the principles of the ADAPT model and evidence-based techniques of modern therapies in the field, including a transdiagnostic approach and the selective use of cognitive behavioural treatment elements such as problem-solving and emotional regulation techniques. The steps outlined in refining the manual are outlined in relation to work amongst West Papuan refugees, and the process of cultural and contextual modifications described during early piloting with Rohingya refugees in Malaysia.
CONCLUSIONS: IAT integrates universal principles of the ADAPT model with the particularities of the culture, history of conflict and living context of each refugee community; this synthesis of knowledge forms the basis for participants gaining insights into their personal patterns of psychosocial adaptation to the refugee experience. Participants then apply evidence-based techniques to improve their capacity to adapt to the serial psychosocial changes they have encountered in their lives as refugees. The overarching goal of IAT is to provide refugees with a coherent framework that assists in making sense of their experiences and their emotional and interpersonal reactions to the challenges they confront within the family and community context. As such, the principles of a general model (ADAPT) are used as a springboard for making concrete, manageable and meaningful life changes at the individual level, a potentially novel approach for psychosocial interventions in the field.
METHODS AND FINDINGS: We conducted a single-blind RCT (October 2017 -May 2019) with Chin (39.3%), Kachin (15.7%), and Rohingya (45%) refugees living in Kuala Lumpur, Malaysia. The trial included 170 participants receiving six 45-minute weekly sessions of IAT (97.6% retention, 4 lost to follow-up) and 161 receiving a multicomponent CBT also involving six 45-minute weekly sessions (96.8% retention, 5 lost to follow-up). Participants (mean age: 30.8 years, SD = 9.6) had experienced and/or witnessed an average 10.1 types (SD = 5.9, range = 1-27) of traumatic events. We applied a single-blind design in which independent assessors of pre- and posttreatment indices were masked in relation to participants' treatment allocation status. Primary outcomes were symptom scores of Post Traumatic Stress Disorder (PTSD), Complex PTSD (CPTSD), Major Depressive Disorder (MDD), the 5 scales of the Adaptive Stress Index (ASI), and a measure of resilience (the Connor-Davidson Resilience Scale [CDRS]). Compared to CBT, an intention-to-treat analysis (n = 331) at 6-week posttreatment follow-up demonstrated greater reductions in the IAT arm for all common mental disorder (CMD) symptoms and ASI domains except for ASI-3 (injustice), as well as increases in the resilience scores. Adjusted average treatment effects assessing the differences in posttreatment scores between IAT and CBT (with baseline scores as covariates) were -0.08 (95% CI: -0.14 to -0.02, p = 0.012) for PTSD, -0.07 (95% CI: -0.14 to -0.01) for CPTSD, -0.07 for MDD (95% CI: -0.13 to -0.01, p = 0.025), 0.16 for CDRS (95% CI: 0.06-0.026, p ≤ 0.001), -0.12 (95% CI: -0.20 to -0.03, p ≤ 0.001) for ASI-1 (safety/security), -0.10 for ASI-2 (traumatic losses; 95% CI: -0.18 to -0.02, p = 0.02), -0.03 for ASI-3 (injustice; (95% CI: -0.11 to 0.06, p = 0.513), -0.12 for ASI-4 (role/identity disruptions; 95% CI: -0.21 to -0.04, p ≤ 0.001), and -0.18 for ASI-5 (existential meaning; 95% CI: -0.19 to -0.05, p ≤ 0.001). Compared to CBT, the IAT group had larger effect sizes for all indices (except for resilience) including PTSD (IAT, d = 0.93 versus CBT, d = 0.87), CPTSD (d = 1.27 versus d = 1.02), MDD (d = 1.4 versus d = 1.11), ASI-1 (d = 1.1 versus d = 0.85), ASI-2 (d = 0.81 versus d = 0.66), ASI-3 (d = 0.49 versus d = 0.42), ASI-4 (d = 0.86 versus d = 0.67), and ASI-5 (d = 0.72 versus d = 0.53). No adverse events were recorded for either therapy. Limitations include a possible allegiance effect (the authors inadvertently conveying disproportionate enthusiasm for IAT in training and supervision), cross-over effects (counsellors applying elements of one therapy in delivering the other), and the brief period of follow-up.
CONCLUSIONS: Compared to CBT, IAT showed superiority in improving mental health symptoms and adaptative stress from baseline to 6-week posttreatment. The differences in scores between IAT and CBT were modest and future studies conducted by independent research teams need to confirm the findings.
TRIAL REGISTRATION: The study is registered under Australian New Zealand Clinical Trials Registry (ANZCTR) (http://www.anzctr.org.au/). The trial registration number is: ACTRN12617001452381.