Cross-sectoral Collaboration in Multidisciplinary Treatment of Trauma-affected Refugees

INTRODUCTION Treatment-seeking trauma-affected refugees possess a complexity involving past trauma and ongoing social stressors, which challenges treatment of mental health problems. There is therefore a great need for developing holistic cross-sectoral interventions, where dealing with these complex challenges are integrated in treatment for trauma-related mental health problems. This research gap has led to the present research study. The overall aim of this study is to examine a psychosocial intervention with an integrated cross-sectoral collaboration for refugees with post-traumatic stress disorder (PTSD) in a randomised controlled trial (RCT) supplemented by a qualitative study. For the RCT the objectives are: 1. To investigate the treatment effect of a psychosocial, cross-sectoral intervention on outcomes of functioning, quality of life and mental health symptoms compared to treatment as usual (TAU) at Competence Center for Transcultural Psychiatry (CTP). 2. To study social stressors as a predictor of severity of mental health symptoms, quality of life and functioning at baseline. 3. To examine predictors for positive outcomes of treatment including social stressors as a predictor. Course of treatment and data collection will follow the SPIRIT statement PATIENT RECRUITMENT Patients can be referred to the clinic by any MD. A senior psychiatrist at CTP assess all referrals, and based on the referral, patients are invited for an initial assessment by an MD at CTP. If it is clear from the referral that the patient does not belong to the clinic's target group, the patient is not invited for an assessment. Previously randomised trials in the CTP have included about 150 patients per year, and the most recent project included 100 patients per year. In all trials 65-75 % of the patients have completed the project. For the present trial the collaborating municipalities cover about 80 % of the referred patients' municipal affiliation. Counteracting this, the investigators expect more patients to be interested in participating in the present trial compared to previous trials that have included psychopharmacological interventions. Therefore approximately 200 project patients are deemed realistic within the given time frame. INITIAL ASSESSMENT The initial assessment is scheduled for all patients that are assessed to be in the target group of CTP. The content of the assessment is not specific for this trial but applies to all initial assessments at CTP. The initial assessment is planned as 2-4 sessions of approximately 45 minutes with an MD, resulting in a total of about two-three hours assessment and consists of; recording of the trauma history, the migration process, social situation, somatic and psychiatric medical history, as well as a clinical and diagnostic assessment. Standardised diagnostic tools such as part of Schedules for Clinical Assessment in Neuropsychiatry (SCAN), the ICD-10 research criteria and the International Trauma Interview (ITI) for ICD-11 section one (PTSD) will be applied in the interview. Various instruments of symptom severity and functioning are completed as self- and observer ratings. Oral and written information about the treatment and the trial is given. If the patient fulfils eligibility criteria and consents to participate, the patient will be randomised after the initial assessment. RANDOMISATION All patients will be randomised after a total of two to three-hour initial assessment with an MD in accordance with inclusion and exclusion criteria. An equal number of patients are randomised in to the two groups (TAU and intervention). The actual randomisation is carried out in REDCap (Research Electronic Data Capture). Stratification by the five municipalities will be carried out before randomisation. BLINDING Blinding patients and practitioners are not assessed to be appropriate due to the different nature of the treatment interventions. However, an intervention-group blinded Hamilton (depression and anxiety) rating will be carried out at the beginning and at the end of treatment. Hamilton-raters will be trained at the clinic and will take part in regular joint ratings to ensure high quality and interrater reliability. Data assessment and data analysis will be performed blinded. REPRESENTATIVITY Patients have not been selected based on more specific criteria than elsewhere in the treatment system and are therefore representative for the population at other clinics treating trauma-related mental health problems in refugees. The results can thus be generalised to other corresponding patient groups and are directly applicable in the planning of treatment for this patient population in general. TRIAL FIDELITY In order to determine trial fidelity, patient attendance will be registered and after each consultation with an MD, psychologist or social counsellor, the topics addressed will be registered, as well as the methods used during the consultation and whether the patient has completed his/her exercises between sessions as planned. At each consultation with an MD any changes in medication will also be recorded. VARIABILITY IN THE COURSE OF TREATMENT All patients will follow the predefined treatment course as accurate as possible, but due to the pragmatic nature of the study there will possibly be some variation in attendance and timing of meetings and sessions, as the patients may become ill or for some other reason not show up for sessions or meetings. FREQUENCY OF RATINGS Patients will be asked to complete self-ratings several times during treatment: at the diagnostic interview/assessment, transition between phase 1 and phase 2, and at the end of the treatment course. Observer ratings and blinded Hamilton ratings will take place at the beginning and end of treatment. In addition, ratings will also be carried out at follow-up 6 months after end of treatment. DATA COLLECTION Data will only be collected at CTP and the municipalities are not involved in any data collection or processing. After each session the clinicians will fill out the case report form (CRF) in the research database through REDCap (Research Electronic Data Capture). All ratings will be carried out by a CTP practitioner (MD, psychologist or social counsellor). Blinded observer Hamilton ratings will be carried out by medical students not linked to the team of practitioners, but to an independent Hamilton rating team with thorough training in using the Hamilton rating scale. In order to ensure quality and inter-rater reliability, the members in the Hamilton team participate regularly in joint ratings under the guidance of one of the senior psychiatrists at CTP (six joint ratings per year). The investigator is overall responsible for the data collection. DATA SECURITY All data collected for this project will be protected according to the General Data Protection Regulation (EU) 2016/679 (GDPR), Act on Processing of Personal Data as well as the Danish Health Act. Information regarding the patients' health concerning trauma history, the migration process, social situation, somatic and psychiatric medical history, medicine, allergies, abuse of drugs or alcohol and ratings will be passed on from the patient record to investigator. SOURCE DATA All data registered about the patients will be kept as source data in the form of original rating forms completed by the patients or practitioners, as well as structured patient records. Data will be saved for 15 years after the trial has ended, which will be stated in a letter of attorney signed by the patients. Case Report Form is source data, this will be described in the source data document, which will be filed in the Trial Master File. QUALITY ASSURANCE Quality control and quality assurance will follow regular procedures as described in sections 3 and 4 of the Danish Executive Order on Good Clinical Practice (GCP). The previous RCTs at CTP has been under GCP monitoring, but the GCP unit has assessed that GCP monitoring is not necessary for the current trial as CTP is assessed to have high-quality internal monitoring. The internal monitoring during the trial follows a manual and is carried out by a team not directly involved in the data collection. The trial is approved by the Danish Data Protection Agency through the Capital Region of Denmark. Managing and filing data will be in accordance with current guidelines for research. Manuals are used in sessions with all clinicians to establish shared standard procedures. In order to ensure interrater reliability, all MDs will attend a SCAN course and regular joint ratings including clinical assessment and observer-ratings will be carried out. POWER CALCULATION A Minimal Clinically Important Difference score for the WHODAS 2.0 has not yet been established. In the literature it has been difficult to find studies with populations comparable to the present study. Based on clinical experience and the sparse available literature a conservative Minimal Clinically Important Difference was taken to be 5 scale points on "WHODAS 2.0 12-item version" and within-groups SD was taken to be 10 scale points. With a power level of 80 % and alpha 0.05 the investigators estimate a sample size of each group of 64 and a total of 128. The completion rate in the preceding randomised trials at CTP was two-thirds, and the investigators therefore set expected drop-out rate to 35 % for this study. Therefore, the investigators increased the number of patients included to 128 x (1/(100%-35%)) and consequently estimated a total sample size of 197 patients. Inclusion will be stopped when approximately 197 patients have been included in the trial. It must be noted that in the case that a Minimal Clinically Important Difference for WHODAS 2.0 is established during the time frame of the trial, it will be considered in the analyses. DROP-OUT ANALYSIS Drop-out analysis is based on the patients who show up at the pre-treatment assessment. The participant will be compared to the patients, who were excluded at the initial assessment on several dimensions in order to identify possible systematic selection bias. The participants included in the trial, but who eventually drop out and do not complete the trial, will be analysed in an intention-to-treat analysis. In addition, completer analyses will be carried out. DATA PROCESSING 1. The primary outcome variables are differences during the treatment course calculated as differences between baseline and end of treatment ratings. The differences between the two intervention groups can be measured with adjustment for baseline and stratification variables by ANCOVA/linear regression and with multiple imputations to handle missing data. 2. The role of social stressors as a predictor for baseline mental health symptom severity, functioning and quality of life will be examined by linear regression with mental health indices, respectively, as dependent variable and social factors as independent variables. 3. Potential predictors of outcome can be analysed by linear regression. A number of analyses are planned with changes in outcome measures as dependent variables and independent variables of social stressors including age, gender and other demographic and baseline data in an attempt to isolate predictors of positive outcome.