2.1 Aims/ Hypotheses
The study aims to characterise the somatic and mental health profiles of individuals with an experience of immigration detention (minimum 28 days) by documenting and assessing participants’ physical and psychological health status upon release from detention at baseline, then at 3, 6, and 12-month intervals and annually thereafter for up to 10 years. Secondly, we will determine whether pre-migration, migration, and post-migration sociodemographic, psychosocial, visa, and clinical factors influence the somatic and mental health profiles of participants who have experienced immigration detention. Lastly, we aim to understand individuals' experiences and attributions of factors related to the impact of immigration detention on their physical and mental health.
We hypothesise that the experience of immigration detention will exacerbate pre-migratory, current, and latent mental disorders despite any intervening treatment and that unfavourable sociodemographic stressors will compound physical and somatic symptoms, with this being corroborated through their narrative experiences as captured by the qualitative data. It is anticipated that the qualitative data will identify the main themes illustrating the somatic and mental health and traumatic experiences of the detained population, whilst the quantitative data will quantify the burden of depression, anxiety, and post-traumatic stress symptoms, in addition to pain, somatic symptoms and functional impairment. Based on limited previous work (Coffey et al., 2010; Sundram et al., 2017), it is predicted that the sample will have high rates of PTSD and major depressive disorder, chronic pain, and increased rates of other chronic illnesses compared with published reports of mainstream population norms. It is further predicted that these rates will decrease with time and treatment in the community (Hocking et al., 2015) and that those with pre-existing mental disorders at baseline will have higher functional impairment over time.
2.2 Study Design/ Theoretical Approach
We aim to understand participants’ unique experiences, observations and attributions of immigration detention and its perceived impact on their physical, mental, and somatic health, as well as pain and functional impairment. The mixed-method prospective cohort design will comprise a quantitative framework with an adjunctive qualitative phenomenological component to facilitate a dual approach to data collection and analysis (Tashakkori & Creswell, 2007; Berman, 2008). Qualitative data will be collected concurrently to provide rich contextual insight into the relationship between immigration detention and physical and mental health. A mixed-method approach enables a comprehensive explanatory framework with both qualitative and quantitative data functioning reciprocally. The quantitative survey instruments are compared with qualitative accounts, which either converge with or diverge from each other in a manner that captures participants' experiences, thoughts, feelings and opinions and respects and preserves their stories.
2.3 Participants and Setting
To be included in the study, the prospective participant must be 18 years or older and have had an experience of post-arrival immigration detention in Australia (either onshore or offshore) for a minimum of 28 days. Individuals whose sole experience of detention is per the Migration Act 1958, s501 ‘character grounds’ act will be excluded on the basis that their detention experience relates to judicial and not immigration matters (Migration Act 1958, subsection 501).
The project will be based in Melbourne, Australia. It will be conducted using in-person, videoconference, and telephone interviews with participants residing anywhere in the country or those who have been resettled in third countries.
2.4 Recruitment Strategy
The target population is known to be traumatised and vulnerable; hence, recruitment needs to be sensitive to these factors. Initially, prospective participants will be approached by their treating team or relevant stakeholders (for example, lawyers and refugee advocates) to determine interest in participating in the study. If they consent, their contact details will be shared with the researchers (BS and DH), who will make contact to discuss the study in detail, enlisting the assistance of an interpreter as required. The prospective participant will be given time to consider their interest in participating, and if they consent to participate in the study, a time for the first interview (T1; baseline) will be scheduled. If a researcher is also a treating clinician, another researcher unknown to the prospective participant will conduct the information-sharing and consent process.
The snowball recruitment method will be concurrently adopted as an adjunctive sampling strategy. This approach is employed to overcome the challenges of recruiting minority or difficult-to-reach communities (Sadler et al., 2010). Snowball sampling will increase the representativeness of the ASR population who have experienced immigration detention in Australia (Magnani et al., 2005).
A voucher worth AUD30 per interview will be provided as compensation for the time spent on the interview. Additionally, the participants will be reimbursed for reasonable expenses incurred as a result of their participation.
2.5 Anticipated risk
Participation in this research endeavour is independent of governmental entities and will not influence the protection claims of prospective participants who may be seeking asylum. Furthermore, participation will not impede their involvement with any services, agencies, or benefits they may access.
If a health condition is identified in the course of recruitment or if a participant becomes distressed during the research interview, the participant will be referred to a health practitioner if consent is provided. Additionally, participants may freely share stories, experiences, perspectives, thoughts and feelings, which may lead to destabilisation. Should this occur, the participant’s treating team will be notified to facilitate a patient review and support or make a referral with their permission. However, research interviews can be a cathartic and healing experience for individuals, providing the opportunity to describe lived experiences (Rosetto, 2014). Previous research noted that whilst follow-up protocols were often time and emotional resource-consuming, participants reported being thankful, reassured of the purpose of their contribution and had an increased sense of empowerment as they reclaimed control over the life narrative and found purpose in their suffering (Matos, 2023). In this case it will be of experiences in immigration detention, often shared for the first time.
If a participant wishes to withdraw from the study, only their de-identified information and data collected until the time of their withdrawal will be kept and included in the research results, and they will cease to be contacted for follow-up interviews. The proportion of participant withdrawals and loss to follow-up will be recorded.
2.6 Study Procedure
2.6.1 Interview protocol
Two researchers (BS and DH) will conduct the research interviews, alternating between facilitation and note-taking. The research team resolved not to audio-record interviews to mitigate the risk of re-traumatization for a population exposed to monitoring and surveillance within immigration detention (Briskman, 2013). Initial baseline interviews will be conducted face-to-face or via video link, considering the participant's preference, privacy, and safety.
At the baseline interview (T1), all participants will be administered the sociodemographic form, five self-report measures, and the MINI (See 2.7). At three months (T2) and six months (T3), the five self-report measures will be re-administered, in addition to receiving an update on any changes to sociodemographic factors in the preceding three months. The latter includes eliciting qualitative narratives of ‘good’ and ‘bad’ life events that have hitherto arisen. At 12 months (T4), the T2/T3 protocol is repeated, and the MINI is re-administered to determine the 12-month prevalence of mental disorders.
2.6.2 Qualitative Data Topic Guide
Guided by the research aim to understand individuals' unique experiences, observations, and attributions of immigration detention and its perceived impact on their physical, mental, and somatic health, as well as pain and functional impairment, semi-structured interviews will initially examine the phenomenology of ASR experiences in immigration detention and their associations with mental distress and somatisation.
In the second step, an iterative topic guide was purposefully refined to target the key themes that arose during the interviews. The topic guide and qualitative data gathered for each topic were then examined using a mixed-methods approach to provide more profound meaning and insight into the quantitative findings. This interview topic guide enables the development of coding labels that will be utilised to develop higher-order themes and will be concurrently and iteratively developed during the interviews. The topic guide is reviewed periodically to maintain relevance until data saturation is attained (Table 2 Appendix).
2.7 Materials
The protocol includes a socio-demographic form and the following instruments:
1) Hopkins Symptom Checklist-25 (HSCL-25) (Mollica et al., 1987)
2) Harvard Trauma Questionnaire for DSM-5 (HTQ-5) (Mollica et al., 2004)
3) Physical Health Questionnaire-15 (PHQ-15) (Kroenke, 2002)
4) Brief Pain Inventory – Short Form (BPI-SF) (Cleeland, 2009)
5) World Health Organisation Disability Assessment Schedule 2.0 (WHODAS; 12-item version, self-report) (World Health Organisation,2010)
6) MINI International Neuropsychiatric Interview 7.0.2 for DSM-5 (MINI) (Lecrubier, 1998; Sheehan et al.,1998)
2.7.1 Socio-demographic form
The sociodemographic form collects information on sex, age, family/marital status, country of origin, ethnicity, language, religion, education level, past and current employment status, migration and immigration detention experiences, and psychosocial indices. (Refer to Appendix 1)
2.7.2 Hopkins Symptom Checklist-25 (HSCL-25)
The HSCL-25 consists of 25 items measuring anxiety symptoms (10 items; Part I) and depression (15 items; Part II). The scale comprises four ordinal categories of response (“Not at all,” “A little,” “Quite a bit,” “Extremely”), rated 1 to 4, respectively. The depression score correlates with major depression, as defined by the DSM-IV (American Psychiatric Association, 1994) (Hesbacher et al., 1980). High concordance (86.7%) has been reported between case assessment by the physician/researcher and the patient's own rating of distress (Hesbacher et al., 1980; Nettelbladt et al., 1993). The HSCL-25 has been translated into numerous languages and validated against gold standard instruments (Kaaya et al., 2003; Sandanger et al., 1999; Sandanger et al., 1998; Silove et al., 2007). The HSCL-25 recommends a cut-off score of 1.75 to indicate clinically significant depression (Mollica et al., 1987; Winokur et al., 1984).
2.7.3 The Harvard Trauma Questionnaire Part IV PTSD (HTQ-5)
The Harvard Trauma Questionnaire (HTQ) is a 4-part cross-cultural screening instrument that was developed as a bilingual and culturally sensitive measure of trauma-related symptoms in refugees (Mollica et al., 1992). The HTQ-5 aligns with the DSM-5 diagnostic criteria for PTSD for use with cross-culturally and linguistically diverse groups (Berthold et al., 2019). Cultural validity and psychometric properties of the instrument have been established in several culturally diverse settings, and the HTQ is the most widely used screening measure for trauma-related symptoms in clinical and research work among ASR worldwide (Morina et al., 2018; Sigvardsdotter et al., 2016). Four ordinal categories of response assess the severity of DSM-5 PTSD symptoms; ‘not at all,’ a little bit,’ ‘quite a bit,’ or ‘extremely,’ rated 1-4, respectively. PTSD symptom severity is computed by averaging the aggregate score of the 25 PTSD symptom items, with 2.50 being the cutoff score for clinically significant PTSD symptoms (Mollica et al., 1992; Wind et al., 2017).
2.7.4 Physical Health Questionnaire-15 (PHQ-15)
The PHQ-15 (Kroenke et al., 1998) is a brief self-report questionnaire comprising 15 somatic symptoms (Kroenke, 2002). The scoring range is 0 (“not bothered at all”), 1 (“bothered a little”) or 2 (“bothered a lot”). PHQ-15 has robust psychometric properties (van Ravesteijn et al., 2009; Zijlema et al., 2013) and has been utilised extensively with refugee and displaced populations (Hijazi., 2012; Hijazi et al., 2014; Morina et al., 2010; Nesterko et al., 2020; Siriwardhana et al., 2013)
2.7.5 Brief Pain Inventory (BPI-SF)
The Brief Pain Inventory – Short Form (BPI-SF) is a patient-reported outcome measure to assess and monitor patients’ pain experiences (Cleeland & Ryan, 1994). The BPI has demonstrated validity and reliability across cultures and languages (Cleeland & Ryan, 1994) and has been utilised in studies of traumatised refugee populations (Jelsness-Jørgensen et al., 2016; Nordin & Perrin, 2019). The BPI-SF contains 11 items that evaluate the severity (4 items) and interference of pain in daily functioning (7 items), which are calculated as separate mean scores (Cleeland, 2008). The interference subscale captures patients’ perceptions of how pain impacts seven constructs related to daily functioning (Cleeland, 2008).
2.7.6 World Health Organisation Disability Assessment Schedule 2.0 (WHODAS 2.0)
The 12-item self-administered WHODAS 2.0 is a short self-report measure for health and disability. It is utilised in both clinical and general populations across a range of health conditions, including mental, neurological, and addiction disorders (Üstün et al., 2010; Tay et al., 2019). The WHODAS 2.0 has sound psychometric properties across cultures and in ASR populations (Tay et al., 2019) and is considered a standardised cross-cultural measure of the functional impact of health conditions across different cultural populations (Habtamu et al., 2017). It assesses the level of difficulty of six domains of functioning: cognition, mobility, self-care, social skills, daily life activities, and participation in community activities. Items are scored from 1 to 5 (none to extreme difficulty), with the aggregate score reflecting the range of functional impairment (WHO, 2010).
2.7.7 MINI International Neuropsychiatric Interview (MINI)
Psychiatric symptom measures will be validated against DSM-5 and ICD-10 criteria using the Mini International Psychiatric Interview (MINI). The MINI was designed as a brief structured diagnostic interview for major psychiatric disorders (Sheehan et al., 1998). The standard MINI assesses the 17 most common mental disorders in clinical and research settings. In epidemiological studies, these disorders were selected based on the current prevalence rates of 0.5% or higher in the general population (Lecrubier, 1998).
2.7.8 Physical health conditions
Where possible, self-reported health status will be corroborated by participants’ health records to ascertain the diagnoses (including onset, duration, and investigations), treatment, and treatment outcomes of health conditions. Physical health conditions will be documented according to the ICD-10 diagnostic categories.
2.8 Data Analysis
2.8.1 Quantitative data
The potential influence of sociodemographic and health-related factors will be analysed using SPSS v22 (IBM Corp, 2013). Hypotheses will be tested by exploring descriptive data and conducting bivariate and multivariate analyses, such as planned comparisons for cross-sectional data and mixed modelling for repeated measures for prospective data. Equivalent non-parametric tests will be used when statistical assumptions are not met. Missing data for self-report scales will be assessed via multiple imputation; aiming for a missing at random (MAR) pattern to ensure that missing values are non-related to unobserved data. Where the MAR is met, imputing up to 20% of the missing data will be considered acceptable.
2.8.2 Qualitative data
We will employ a thematic analysis of the data from the interview transcripts (Braun & Clarke, 2006) using the QSR NVivo 10 software (QSR International Ltd, Melbourne, Australia, 2014). The themes will be identified in an inductive fashion, which is an appropriate method for research in which there exist nil or few previous studies (Elo & Kyngas, 2008).
The transcripts will be analysed in four distinct phases (Braun, 2006; Lincoln & Guba, 1985; Levitt et al., 2021). The first involves generating initial codes from the data to organise them into meaningful groups. Participants’ perspectives and thoughts in response to the research questions will be selected and highlighted to inform the development of a coding template that supports the respective categories. The initial codes are collated into broader overarching themes in the second phase.
The third phase involves reviewing themes and assessing congruity, confirmatory data, and/or unique insights. Here, qualitative material in the form of quotes and statements is linked with and provided to support and elucidate the significance of each theme. The final phase involves finalising the emerging themes in the context of the research questions relating to the impact of the immigration detention experience on somatic and mental health outcomes in the interviewed ASR population. At this final stage, the second author will cross-check the themes independently and later corroborate them within the research team to enable inter-coder reliability and consensus (O’Connor, 2020). A standardised approach to conducting and reporting qualitative data will be employed using the Standards for Reporting Qualitative Research (SRQR) protocol (O’Brien et al., 2014).
2.8.3 Mixed Methods Analysis
Mixed-methods research, or the collection and analysis of both quantitative and qualitative data and its integration, employs the strengths of both approaches and allows for a deeper understanding of a phenomenon of interest, especially when the phenomenon is complex (Howe., 2012; Guetterman et al., 2015; Creswell 2011 and 2015). The mixed-methods approach is increasingly utilised and is recommended when findings from a research approach can better be interpreted by the findings of another approach or when quantitative findings can be enriched by qualitative findings (Creswell, 2015). The mixed method facilitates convergence and divergence of qualitative and quantitative data and results in a more holistic picture and response to the research question (Santos et al., 2017). A triangulation strategy will be adopted with qualitative and quantitative data given equal weight. The integration and synthesis of the data will occur in a matrix, where quantitative findings are subsequently considered alongside qualitative insights to enable and produce greater meanings and observations.