Received: 11 February 2005
Accepted: 18 April 2005
This paper aims to describe culturally sensitive rather than culturally specific mental health practice in East London, and to show how this can be achieved with a small team. It tracks the development of a specialised service for a diverse population of refugees with a diagnosis of post-traumatic stress disorder since 2000. Demographic information for the first 432 patients illustrates their wide ethnic, linguistic and psychological backgrounds and needs. Cultural sensitivity has assisted patients in engagement with Western psychological treatments. Attendance rates are as good for refugee patients as for non-refugees, and for non-English speakers as for English speakers. Cognitive– behavioural techniques are employed which include groundwork to establish trust, direct exposure to traumatic memory, and multi-agency rehabilitation to increase social engagement and prevent relapse. Outcomes of treatment are being evaluated but cultural issues in diverse and changing communities have to be addressed pragmatically by clinicians working collaboratively with their refugee patients.
cognitive–behavioural therapy, cultural sensitivity, post-traumatic stress disorder, refugees
Refugees have significant and complex mental health needs. For clarity, the term ‘refugee’ here fulfils the United Nations (UN) criteria of a person who:
owing to a well founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his [sic] nationality and is unable, or owing to such fear, is unwilling to avail himself of the protection of that country ... (UN, 1951)
This definition is much wider than that used by the British government, which restricts the term ‘refugee’ to those whose request for asylum has been granted, but is that most commonly used internationally and among practitioners.
The prevalence of post-traumatic stress disorder (PTSD) in refugees ranges from 4% to 86%, and for depression from 5% to 31% (Hollifield et al, 2002). Those who have been exposed to high numbers of war-related traumatic events have a high incidence of PTSD (60.5% in one study by Ai et al, 2002), and comorbid disorders such as depression and anxiety. There is also evidence that refugees diagnosed with PTSD have a lower subjective quality of life than other patient groups suffering from severe depression, psychosis and alcoholism (d’Ardenne et al, 2005).
Forced migration and acculturation may compound the effects of the initial trauma and increase the risk of PTSD (Summerfield and Toser, 1991; Mollica et al, 1993). Detention of asylum-seeking refugees has led to serious subsequent mental health problems for some (Pourgourides et al, 1997). Refugees with a wide range of health needs (or, more strictly in view of UK government terminology, asylum seekers awaiting determination of their status) may be dispersed around the UK by NASS (the National Asylum Support Services: www.homeoffice.gov.uk) at very short notice, sometimes with only hours to pack and inform health service providers. Health assessment, treatment and planning then have to be transferred abruptly to other NHS trusts across the country.
Psychological treatments for refugees have been varied, with little published evidence on treatment outcome. They include psychodynamic methods (Bustos, 1992), testimony method (Cienfuegos and Monelli, 1983), cognitive therapy (Foa and Meadows, 1997) and cognitive–behavioural interventions (Basoglu, 1992, 2004; Paunovic and Oust, 2001). Drozdek (1997) has outcome data on the effectiveness of group psychotherapy with Bosnian refugees in Holland, showing refugees to be responsive to therapy that integrates their traumatic memories and affect. Yule (2000) has recommended group treatment for refugees who have been exposed to similar life-threatening events. De Jong et al (2000) propose psychosocial interventions, which focus on strengthening a whole community through psycho-education. Overall, Van der Veer (1998) has described how cultural competence and sensitivity can inform the adaptation of Western cognitive–behavioural therapy (CBT) models to meet the needs of refugees in Western clinical settings.
The NHS has a legal and moral requirement to respond to the health needs of refugees in a culturally appropriate and competent way. The National Institute for Clinical Excellence (NICE, 2005) includes refugees in its evaluation of the treatment of PTSD. It recommends trauma-focused CBT, which includes imaginal and in vivo exposure to traumatic memories and stimuli, as treatment of choice. It further recommends that health professionals familiarise themselves with a sufferer’s cultural background (NICE, 2004, Paragraph 18.104.22.168.6.), and that ‘differences of culture and language should not be seen as an obstacle to the provision of effective trauma-focused psychological interventions’ (Paragraph 22.214.171.124.4.).
Refugees often lack understanding about the NHS and how it works, have communication issues, and frequently have to meet professionals who lack awareness of asylum issues and cultural competence (Quickfall, 2004). There is some evidence that older refugees with PTSD have increased resistance to learning English and that this might impact upon their acculturation and engagement with psychotherapy (Drozdek, 1997).
In addition, there remains an apparent lack of commitment from national and local health services to provide appropriate and accessible help for refugees (Burnett and Peel, 2001; Mahtani, 2003). This situation may reflect clinicians’ fears that they lack the necessary cultural skills and knowledge needed to form therapeutic and supportive relationships with refugees.
London’s eastern boroughs provide a microcosm of the world, its transitions and its mental health problems. The Bangladeshi community of Tower Hamlets is well known. In Hackney and Newham more than 50 minority ethnic communities make up over 40% of the population of 700 000, a mixture that is ever changing and which contains a significant number of refugees from global and regional conflicts.
The Institute of Psychotrauma is part of the East London and City Mental Health NHS Trust, which was established in 2000 to provide local boroughs with appropriate and accessible mental healthcare, and with a strong policy commitment to making those services local and accessible (www.elcmht.nhs.uk). Prior to 2000, refugees requiring specialist PTSD services were sent to other centres but these included clinics which explicitly excluded non-English-speaking patients. Since April 2000, East London residents have been referred to the Institute if they:
• suffered from serious post-traumatic psychological disorders
• were supported within secondary care.
The institute retains a small team with three part-time clinicians, their trainees, and a researcher, all of whom are first- or second-generation immigrants speaking two or more languages. All have local knowledge and also work in East London community mental health teams. The team has non-UK perspectives and values that members believe increases rapport with people who are themselves in transition. The team holds an ethos of cultural relativity in all its work, and places a high value on respect for and curiosity about all new cultures. Since 2000, the Institute has accepted 432 referrals for assessment and/or treatment, of which 207 were self-defined as refugees. Refugees constituted 28% of the referrals in 2001. By 2004 this had risen to 48% of the total patient list.
This paper provides a descriptive account of the service based on audit data that are routinely collected from referrers contacting the clinic and stored on an encrypted database.
The 208 refugee patients defined their ethnicity as follows: Black African, 25%; Albanian/Kosovan, 22%; Kurdish, 14%; Turkish, 10%; Algerian, 7%; Sri Lankan, 4%; Iraqi and Iranian, 3%; and ‘other’, 12%. Seventy percent of refugee patients required language support. The origins of the trauma for refugees were as follows: 41.3% were victims of war; 44.2% were victims of torture; 11.1% had crime-related trauma; 1% were traumatised as a result of an accident; 2.4% had other minor sources of traumas. The profile for nonrefugees was quite different. Fifty-three percent were victims of crime; 17% were victims of accidents; 13% the victims of torture and 11%were victims ofwarfare. Categorisation of trauma was subjective. A life events checklist was used.
The service excludes people who are receiving psychological therapy elsewhere, those under the age of 18, and those who are actively psychotic. It is made clear to all patients, at initial contact, that written reports are available to them and their referrers only and that the institute does not prepare reports for outside agencies such as immigration services. The clinicians in the service emphasise that they are there to form an alliance with the patient and not to prepare reports regarding the legitimacy of a claim for asylum.
Following referral, patients are given an ‘opt in’ letter in their first language, as well as literature that describes the purpose, ethos and location of the service. The rationale is to provide patients with a clear choice and a sense of ownership of a process, when so much of their refugee experience has entailed little information and little choice. Once patients have opted in, they are given a first assessment appointment.
Below is some evidence for engagement with the trauma service. Attendance of refugee and non-refugee patients has been compared as well as attendance of those requiring language support and those able to communicate without interpreting services.
Table 2 shows data about 91 patients who were selected as appropriate for the service but did not attend any sessions (21% of the total number of referrals). There was little difference between the percentages of those who were defined as refugees and those who were not. Similarly, there was little difference between the proportion of non-attenders who were known to have language support needs and those who could speak English.
Table 3 provides data on the remaining 337 users who attended the service at least once. An independent t-test revealed that there is no statistical difference between thenumbers of appointments attended by the two groups (P = 0.60).
Table 4 shows the mean number of those who failed to keep appointments (‘did-not-attend’ or DNAs) and excludes cancellations. DNA figures for refugees were actually lower than those for non-refugees, although the difference was not statistically significant (P = 0.31). English-speaking patients had a higher rate of DNAs than those who required language support (P = 0.003). Patients who required language support attended fewer appointments (mean = 4.78) than English speakers (mean = 5.76), and the difference approached levels of statistical significance (P = 0.057). The reason for the smaller number of appointments may be that interpreted sessions were longer and that therapy therefore occurred over fewer sessions that were of greater duration.
The current DNA rate for all appointments is 21%, which is comparable with the Trust’s own audited rates (www.elcmht.nhs.uk). After two DNAs, patients are discharged from the service, although efforts are made to audit why and how they are lost to the service, particularly after they have made an initial attendance. Some refugees may have been dispersed and thus not received the letter at their new address (which will almost certainly be outside the catchment area for this service). Refugees who still reside in East London often have insecure addresses shared by many in bed and breakfast (basically insecure shared rented) accommodation, with no individual post box or landline telephone. In these settings letters may be lost or intercepted. A proportion (15%) of these refugees have already begun to receive psychological help from other sources. Others decide that they now wish to be managed within their immediate community or to cope without help at all. Some were in the middle of major transition such as having a baby or receiving medical or surgical treatment, and therefore decided to postpone psychological treatment. Some patients were fighting their legal appeal/facing deportation/ repatriation, and were more concerned about ensuring that their reports were in. Consequently they wanted to delay or cancel psychotherapy.
At a more personal level, patients fear being emotionally overwhelmed through revisiting and aggravating their PTSD symptoms. They fear repatriation most of all (Ai et al, 2002). They may be mistrustful of any government agency and not appreciate the meaning of clinical confidentiality (Burnett and Fassil, 2002). Van der Veer (1998) cites the reluctance of attendees with PTSD who fear re-exposure to their traumatic pasts coupled with feelings of embarrassment and shame. The British Medical Association (BMA, 2002) describes the lack of translated information about specialist services, leaving refugees unaware of the range of services available and their rights of access to them.
The communication and language support needs of all patients with refugee experience are ascertained at the time of referral. Patients’ letters have been translated into the six most widely used languages: Turkish, Albanian, Arabic, Somali, French and Bengali. There is ongoing work to access a much wider and more flexible translation of materials to support the service. Efforts may also have to be made, via the referrer, to establish that there will be somebody who can translate the documents that refugees receive in the post from this service. Clinicians’ fluency in another language allows them to understand the limits of ‘linguistic equivalence’. Their languages may not be the same as that of the refugees, but there is an awareness of the lack of correspondence of meaning between one language and another (d’Ardenne and Mahtani, 1999; BMA, 2002).
All clinicians are experienced in using interpreters and advocates. The service uses professional interpreters who have experience in interpreting the traumatic histories provided by refugees from areas of conflict. The institute has access to nearly 50 languages from a local authority interpreting service. Many of the interpreters live in East London and have local knowledge of community services and networks. Some may also have had refugee experience.
Whenever possible the patient is given choice about the language to be used during assessment and treatment. Offering choice empowers the patient and gives individuals more of a voice. The therapist model is essentially a collaborative one with the interpreter and with the patient. The clinician starts froma position of seeking to understand the culture of the patient, acknowledging his or her own lack of objectivity, and the social and structural inequalities that exist between the patient and the psychotherapist (Patel, 2003). Care is taken to ensure gender or ethnic sensitivities are addressed, and that interpreting does not generate any further barrier for the patient. This would certainly be the case for refugee patients with a history of sexual torture or rape, but it might be equally undesirable, for example, to have a Turkish national interpreter for a Kurdish patient. Occasionally, a suitable interpreter cannot be found, perhaps because the language concerned is not widely spoken in the locality. Sometimes a patient turns up unexpectedly and no interpreter is available. In these circumstances the institute uses a telephone interpreting service, which can be accessed quickly. This back-up service provides flexibility and anonymity, although it is more costly than regular face-to-face interpreting.
Patients occasionally bring their own community advocates and/or family members who provide good support in the waiting room, but family members are never used as interpreters. The patient is encouraged to rely on the clinician and interpreter for more detailed trauma assessment and treatment sessions. Every effort is made to locate an interpreter who may already be known to and trusted by the patient from previous clinical appointments, but who is not an immediate member of the community.
All patients are comprehensively assessed with standardised psychometric tools, to provide an overall measure of the impact of the service before and after contact and, where possible, to make limited comparisons with other psychiatric patient groups (d’Ardenne et al, 2005). Assessment includes the Manchester Short Assessment of Quality of Life (MANSA, Priebe et al, 1999), the Impact of Events Scale (IES, Horowitz et al, 1979), the Beck Depression Inventory (BDI, Beck and Steer, 1987) and the Beck Anxiety Inventory (BAI, Beck and Steer, 1988). During the first appointment, clinicians assist patients who have been unable to complete the questionnaires. Thereafter, clinicians work to a person-centred and culturally sensitive assessment for each patient with refugee experience. For example, they seek updated information about the regional conflicts that refugees have fled, to prevent oversimplification or stereotyping of their difficulties. This takes time and assessments may take two or three visits before completion. Atlases, paper and electronic, as well as the internet are used collaboratively with patients to establish a clear account of their history and geography. Both the patient and the clinician, as shared learning, keep internet print-offs. Local, national and international resources are collaboratively researched as required. Patients are provided with detailed feedback of their assessments, with full attention to these cultural and social contexts, and then offered a choice of interventions provided by the service.
If the clinician and patient are in agreement, an intervention of trauma-focused CBT is undertaken. The average number of sessions is five, with a range from 2 to 30. However, several factors, such as having to attend numerous hospital appointments, lack of childcare or high levels of distress, can affect the frequency and duration of contact. Consequently, some patients are seen either fortnightly or monthly.
The institute uses a model of engagement proposed by Herman (1992), which is appropriate for patients such as refugees who have been exposed to multiple stressors over a period of time. Treatment consists of three stages: these include the establishment of safety, direct psychological intervention and finally reintegration into the community. The programmes for refugees differ in respect of the extent of the groundwork that needs to be done to establish trust before any revisiting of the traumatic memory can be considered.
Phases of this groundwork include increasing the refugee’s social support; teaching self-care; emotional regulation; and connections with aid agencies, all of which are done from a culturally sensitive perspective. Clinicians may liaise with refugee agencies with family tracing facilities, for example the International Red Cross (www.icrc.org), Amnesty International (www. amnesty.org.uk) the Medical Foundation for the Care of Victims of Torture (www.torturecare.org.uk), and the Refugee Council (www.refugeecouncil.org.uk). Patients may also require CBT for help with dissociative episodes, including flashbacks, and distressing symptoms such as nightmares, poor sleep, anxiety or phobias.
The treatments selected as most appropriate for PTSD symptoms are:
• reliving the traumatic memory
• the identification and targeting of ‘hot spots’ – that is, the most distressing moment within a trauma memory and associated negative appraisals (Grey et al, 2002)
• managing the subsequent distress
• incorporation of new and incompatible cognitions about the traumatic event that restructure the traumatic memory.
Exposure of the traumatic memory has been shown to be particularly effective for reducing fear and anxiety, and all the avoidant symptoms that emanate from PTSD (NICE, 2004). However, there are other distressing emotions. When shame, guilt or anger is related to the trauma, for example in rape (Herman, 1992), cognitive restructuring is used. In this technique, the meaning of the traumatic memory is modified outside the reliving exercise. The patient is asked to rehearse these new appraisals of the trauma, which are incompatible with the old memory. These new appraisals are then inserted back into the relived memory. For example, ‘I am dying’ or ‘I could be harmed’ become ‘I survived’ or ‘I escaped’.
Reintegration is aimed at increasing social engagement and prevention of relapse, and represents the final stage of the intervention. The institute makes detailed recommendations both to the patient and to the referrer about how this might be achieved. Examples of this include:
• supporting the patient within existing social and family networks
• continued assistance in helping the patient trace lost family members and friends both in the UK and overseas
• extending the patient’s role within educational or occupational development – including language and vocational training
• engaging the patient with local community groups campaigning for political and social justice for their refugee communities
• dealing with housing, immigration and legal problems quickly and effectively through refugee and local services; addressing other healthcare needs
developing recreational, sport and cultural activity appropriate to the patient’s needs and background.
Burnett and Peel (2001b), Kleijn et al (2001), Mahtani (2003) and many others have argued that Western psychological models cannot be used on non-Western populations. Burnett and Peel (2001b) and Summerfield (1991; 2001) have argued that PTSD is itself aWestern construct and is inappropriate for refugees on a number of indices. They argue that refugees are in a state of acute distress and that their distress should not be pathologised or medicalised. The difficulty with this position is that it leaves the clinician with very few options, and with distressed patients who have no service. The institute receives patients with complex and severe problems. Each patient is viewed in a person-centred way. DSM-IV and ICD 10 standards are used for guidance only and never as standalone criteria. Clinicians draw on the intrapersonal, interpersonal, cultural and political contexts of the patient to make a formulation, on which treatments are based.
There remains a fundamental debate in mental healthcare for refugees about the need for cultural specificity versus the need for cultural sensitivity in treating diverse populations (Bhui and Olajide, 1999; Bhui and Sashidharan, 2003). In East London there are practical reasons for a non-specific, culturally sensitive approach to mental health problems, especially for the needs of refugees and particularly those suffering with PTSD. Referrals have come from over 27 different countries in the past four years; the institute has hired interpreters for 18 languages in that time. Even if ethnic matching of therapist to patient were appropriate, it would be impossible within staff resources to do so.
There are other issues. Small communities are often split. Individuals may not wish to see mental health professionals from their own background for fear of espionage or poor confidentiality. Some patients fear stigma, for example the shame of rape. Some patients re-experience traumatic memories while being interviewed by members of their own community. Therefore, a more pragmatic model of working transculturally has evolved (d’Ardenne and Mahtani, 1999).
Patel (2003) has argued that assessments, formulations and recommendations must extend beyond the purely psychotherapeutic, and reach to the cultural and social contexts of patients with refugee experience. D’Ardenne et al (2005) looked at the outcome of 117 patients in this service, of whom 28% were refugees. Preliminary data show that refugees are more symptomatic than non-refugees, with high levels of anxiety and depression and remarkably low subjective quality of life (SQOL). Low mood was determined not just by traumatic events, but also by dissatisfaction with family, friends, employment possibilities, finances, mental health, and life in general. Patients with PTSD have lower SQOL than other psychiatric groups as measured by every domain with the exception of mental health. High levels of depression and anxiety, being older, and being from a minority ethnic group predicted low SQOL.
Mainstream mental health services must take responsibility for ensuring that both the workforce and its practices are sensitive to difference and value it. Specialist, separate development lets the dominant culture ‘off the hook’ in its responsibilities to a wider community. It places culture or race above all other considerations of difference, and pathologises its role in the genesis of PTSD. The Institute of Psychotrauma is taking important steps to provide a service for traumatised refugees in East London. Clinicians require a range of cross-cultural competencies for work with refugees (d’Ardenne andMahtani, 1999; Quickfall, 2004; Box 1). There are no ideal solutions to these complex issues faced every day in the NHS. However, culturally sensitive rather than culturally specific services are the most pragmatic solution in East London where there are large, significantly diverse and everchanging populations.
Clinical practice in PTSD with refugees continues to evolve, and places cross-cultural demands on patients and clinicians (www.nice.org.uk). If attention is given to engagement and the establishment of safety within therapy, take-up of the service is no more difficult for refugees or for those requiring interpreters. This service can undertake brief cognitive work aimed at revisiting patients’ traumatic memories. Its policies favour inclusion of refugees and those requiring language support. Evidence in this paper has shown that attendance is no worse, and that DNA rates are actually better, than for non-refugees, although the 21% of accepted patients who never attend provide no grounds for complacency.
|Box 1 Components of cultural competence (after Quickfall, 2004)|
|Institutional regard||Institute’s policies on inclusion for diverse cultures and languages in|
|Access to services||50% referrals from ethnic minority; sensitivity to sex and culture of
therapist and interpreters
|Cultural communication||Comprehensive assessment of cultural and linguistic needs at point of
|Cultural insight||All therapists trained in transcultural approaches and use of interpreters|
|Cultural understanding||Provision of comprehensive cultural advice to secondary services for
refugees within the trust
• NASS: National Asylum Support Services www. homeoffice.gov.uk (accessed 20 April 2005).
• Amnesty International www.amnesty.org.uk (accessed 20 April 2005).
• East London and the City Mental Health NHS Trust www.elcmht.nhs.uk (accessed 20 April 2005).
• International Federation of Red Cross and Red Crescent Societies www.ifrc.org (accessed 20 April 2005).
• National Institute for Clinical Excellence www. nice.org.uk (accessed 20 April 2005).
• Refugee Council Online www.refugeecouncil.org•uk (accessed 20 April 2005).
• Medical Foundation for the Care of Victims of Torture www.torturecare.org.uk