Challenges of Combining Perspectives.

BACKGROUND: Asylum seekers have increased risk of suicide and suicidal behavior, with differences related to origin, gender, and age. There are barriers to communication in clinical encounters between asylum seekers and clinicians. There is insufficient knowledge about how communication in the clinical encounter affects the suicide risk in female asylum seekers. AIMS: To explore the documented communication between female asylum-seeking suicide attempters and clinicians and how it affects treatment.
METHOD: The medical records of 18 asylum-seeking women who had attempted suicide were analyzed with content analysis.
RESULTS: Communication between patients and clinicians was affected by: the unbearable realities of the women; difficulties for clinicians in decoding languages of distress, and understanding trauma and subjective meanings of suicide; challenges of combining patients' and clinicians' perspectives; and a sense of shared powerlessness. LIMITATIONS: The medical records did not give direct access to the patient's experience, only to the patient as documented by the clinician.
CONCLUSION: The results suggest that clinicians working with asylum seekers who have attempted suicide need to develop an understanding of social and cultural factors and of trauma issues. A question for further study is how an enhanced integration of context and subjectivity in psychiatric practice would equip clinicians for the specific challenges encountered.

Asylum seekers are individuals who have applied for protection as refugees in a foreign country but have not yet been granted asylum. Asylum seekers are vulnerable to mental ill-health, but often have difficulties accessing health care and receive lower-grade treatment than comparison groups in the population (Goosen et al., 2011; Joshi & Warfa, 2015; Kirmayer et al., 2011; Maier, Schmidt, & Mueller, 2010).

Studies of asylum seekers have shown increased rates of attempted and completed suicide compared with the general population, with differences in risk related to origin, gender, and age (Goosen et al., 2011; Kalt, Hossain, Kiss, & Zimmerman, 2013; Staehr & Munk-Andersen, 2008). The suicide rates of immigrants in general tend to correspond with the suicide rate in their country of origin (Ratkowska & De Leo, 2013; Spallek et al., 2015). Studies have demonstrated high rates of attempted suicide for immigrants depending on origin and gender, with particularly high rates for some groups of female immigrants (Bhugra et al., 2011; Bhui, McKenzie, & Rasul, 2007; Bursztein Lipsicas et al., 2012; Bursztein Lipsicas et al., 2013; Kosidou et al., 2012).

In clinical encounters, the communication between clinicians and asylum seekers is sometimes complicated by language barriers and cultural, migration-related, and legal factors. There is a growing awareness of cultural idioms of distress, defined as variations in culturally and socially shared ways of expressing mental suffering (American Psychiatric Association, 2013). In a European Psychiatric Association (EPA) guidance document, the need for clinicians to have cultural competence is stressed. Cultural competence is defined as the ability to be aware of cultural factors in the therapeutic interventions and to reflect on the impact on the interaction with the patient of the clinician's own cultural identity and prejudices (Schouler-Ocak et al., 2015). Cultural competence should be considered both at the individual and the institutional level. At the institutional level, access and availability to care, the use of qualified interpreters or culture brokers, and culturally appropriate structures such as food and room for prayer should be provided (Schouler-Ocak et al., 2015).

Stress disorders, including acute stress reaction and posttraumatic stress disorder (PTSD), have been shown to predict suicide (Gradus, 2017). Exploring suicidality in the clinical encounter with refugees is not only complicated by cultural and language barriers but also often by severe stress reactions, for example, cognitive impairment related to experiences of torture and shame related to experiences of sexual violence (Crumlish & Bracken, 2011; Kirmayer et al., 2011).

Kirmayer et al. (2007) and Colucci, Lester, Hjelmeland, and Park (2013) recommended exploring the subjective meaning of suicide, including cultural variations in meaning. Michel, Dey, Stadler, and Valach (2004) emphasized the need for a patient-focused approach in which the patient's narrative is elicited. To understand factors that influence the clinical assessment and treatment of asylum seekers who have attempted suicide, we in an earlier study analyzed the medical records of 88 such men and women (Sundvall, Tidemalm, Titelman, Runeson, & Bäärnhielm, 2015). The results indicated that the female asylum seekers suffered from more serious mental health conditions than female suicide attempters who were permanent residents with whom they were compared: The female asylum seekers received compulsory care more often, had longer periods of hospitalization, and received a greater variation of pharmacological and other forms of treatment. Yet, the asylum seekers, both men and women, were referred to specialized follow-up less often than the patients who were permanent residents. We concluded that there was insufficient knowledge about how communication in the clinical encounter affects the assessment and management of suicide risk in asylum seekers in general and female asylum seekers in particular.

The aim of this study was to explore the communication documented in the medical records between female asylum-seeking suicide attempters and clinicians and how it affects the clinical encounter and treatment.

Method

Subjects

Of 88 asylum-seeking patients identified from the medical records of patients registered for suicide attempts from 2005 to 2009 at the Stockholm County mental health central emergency services, all 18 women who had received compulsory care were selected. The studied patients were 19–43 years old, 11 were younger than 30. Seven patients were born in Bangladesh, two in Jordan, two in Azerbaijan, and one in each of Uzbekistan, Iran, Afghanistan, Iraq, Pakistan, and Congo. They had resided in Sweden between a few months and more than 5 years, but in the majority of cases for less than 2.5 years. One of the women was sent directly to a hospital in another county after initial assessment. The others were hospitalized for between 5 and 197 days, the majority for less than 2 months. Of the 18 patients, four were followed up in primary care and 12 in specialized mental health services; in two cases follow-up was not documented.

Information on the use of interpreters was fragmentary in the records. However, at the time, interpreters were available at the emergency services, even though their qualifications varied. Culture brokers were not available. Cross-cultural consultation and supervision were not well-known to the staff and rarely used in the studied cases.

Material

We studied the digitalized medical records from the initial assessment in the emergency services and from the hospital ward if the patient was hospitalized in the same hospital. Similarly, outpatient care records were available for our review when the care was provided within the same mental health services as the emergency care unit, and from the emergency services and hospital wards if the patient was re-admitted over time.

Medical records are written by all professions (physicians, nurses, social workers, psychologists, mental health workers, occupational therapists, physiotherapists) in the services, and updated after each assessment, treatment intervention, conversation with the patient, communication with other services, network etc. Normally there are several items added every day, of varying length. Formal suicide assessments are made by psychiatrists. Specific categories of suicidality or suicidal behavior are recommended, but at the time of the study, using them was not mandatory. Information on suicidality could thus be found in all types of records.

Data Analysis

The medical records were analyzed with content analysis (Krippendorff, 2004), using the software NVivo9 (Bazeley & Jackson, 2013). The following procedure was applied: (a) codes were assigned to meaning units gleaned from the texts; (b) codes were grouped into preliminary categories; (c) broader themes were identified with the help of written memos synthesizing the findings of each record; (d) potentially relevant codes for each theme were further explored and the themes were refined.

One author (MS) coded all records after which the codes were discussed and consensually adjusted in discussions with another author (SB). All authors read successive drafts of themes. In the end, again by consensus, six themes were formulated.

Ethical approval was given by the Regional Ethical Review Board of Stockholm (number 2010/3:6 and 2012/982-32).

Results

The analysis yielded the following overriding themes: unbearable realities (referring to the communication on the situation in the asylum process and the situation as a woman); decoding languages of distress (referring to the patients' difficulties in communicating distress and life circumstances, and the clinicians' difficulties in decoding unfamiliar expressions and circumstances); elusive trauma (on how trauma was communicated, documented, and taken account of); meanings of suicide and communication of meaning (on how the patient's subjective meaning of the suicide attempt was documented and taken account of); challenges of combining perspectives (on how clinicians and patients communicated on different perspectives on the patient's situation and health state); and shared powerlessness (on documented expressions of powerlessness in the patient and in the clinicians).

Unbearable Realities

The first theme focuses on the documentation of the communication about external realities that both patients and staff experienced as extremely difficult to handle. Two categories within this theme were salient: the asylum process, and the situation as a woman.

The Asylum Process

The women often communicated distress related to the asylum process. Particularly distressing aspects were negative decisions in the asylum process but sometimes even more so the long waiting time and insecurity, and difficulties understanding the roles of different authorities. One woman was reported to express that, "no to asylum is equal to a death sentence." Another group of stress factors included the fear of having to return to the home country, worries about relatives, housing and financial problems, and weak social networks.

Clinicians communicated stress and frustration related both to difficulties they experienced assessing the mental health in the asylum situation and to what they felt were pressing external factors, beyond the medical considerations. There were examples of staff communicating difficulties in understanding and being irritated by the actions of the migration authorities, as well as compassion with patients facing pending or immediate deportation.

The Situation as a Woman

Of the 15 women who had been exposed to trauma, nine were sexually abused. For five women the abuse was part of torture in prison in their home country; for two others it had taken place at work, either in the home country or in Sweden. One woman was abused by a man she hoped to marry, and the abuse another woman had experienced was related to social misery and prostitution.

In the women's communication the social consequences of abuse were described as devastating for cultural reasons: Friends, neighbors, and relatives frequently condemned the abused woman and broke all relations with her. Some patients were afraid of losing their current housing as lodgers with persons from their country of origin, should the truth be disclosed. "Nothing worse can happen to a woman in her lifetime," one woman explained – she was stigmatized in her own eyes, too. "Women that are raped and single have no value," she added.

The inner psychological consequences of sexual abuse were described as "an open wound" or "living torment." One woman described the loss of identity inflicted by trauma. After having fought for women's rights and liberties, she now felt "crushed."

Two women who had had extramarital relations were condemned by relatives, and said that returning to the home country would be a danger to their lives. Even though, for one of them, the relationship to a certain man had led to repeated rape, her only hope was for the offender to marry her in order to "transfer the shame" onto him.

Several women also described a lack of opportunities for girls, generally oppressive conditions, physical abuse, and forced marriage in their families of origin, as well as domestic violence and gender-based subjugation in marriage. In one case, the husband's attitude to the adolescent daughters, which had led to the split of the family, contributed to the mental ill-health of the woman.

Of the 13 women who were mothers, 11 had all of their children in Sweden. For most of them the pain of not being able to fulfill the responsibility as a mother was indicated as unbearable.

Clinicians communicated that it was difficult to grasp – or even to believe in – the extent of the ordeals communicated by the women. Dialogues or therapeutic interventions aiming to help women to handle their realities were rarely documented. When children were involved, the clinicians usually communicated a concern for their well-being.

Decoding Languages of Distress

This theme covers the documentation of how clinicians struggled to decode a variety of languages of distress. The theme can be divided into four categories: difficulties of contact and verbal communication; unfamiliar behavior and symptoms; unfamiliar forms of expression; unfamiliar life circumstances.

Difficulties of Contact and Verbal Communication

Noticeable barriers to communication were the patient not giving any contact, being silent, being incoherent, having extreme language difficulties, or being illiterate. Relatives and interpreters sometimes described a patient's speech as meaningless or impossible to translate.

The following note in the records seems to reflect the clinician's frustration when a patient was seen to avoid direct communication: "Repeated attempts are made to get information on X's asylum application. X then interrupts, saying ‘Don't disturb me, I have a headache'." The clinicians sometimes interpreted the patient's lack of communication as a difficulty in talking about trauma. Reasons given for this were that it was too painful; shame and guilt feelings; using evasion and forgetting as coping; high levels of tension and fear; language difficulties; memory problems; and the expectation that the staff would not understand.

Fluctuations in relating and lack of verbal communication were often noted as lack of compliance. Only rarely did the clinicians reflect on how the presence of interpreters influenced the fluctuations. If the contact with a patient was easier in the presence of one or more family members, this sometimes led to the perception of the patient as less credible.

Unfamiliar Behaviors and Symptoms

There were records of women falling to the ground, changing from mutism and lack of psychomotor activity to intense anxiety and confusion or aggression, moaning and rocking their body, or screaming in an uncontrolled way. The meanings of these symptoms were rarely explored. One record stated: "Bangs her head on the wall. Since she is angry, this [behavior] is assessed as acting out." One woman's attacks of vertigo, double vision, nausea, and headache were interpreted as confusion or psychosis. Later, after her husband described earlier such attacks as stress-related, the attacks were interpreted as dissociative symptoms.

Patients' refusal to eat and drink was seen either as a sign of depression, a crisis symptom, or as a conscious protest against the life situation or against health care.

Unfamiliar behavior, especially combined with agitation, was sometimes described as "hysteria" in the records. Hysteria was used in the everyday sense of the word, to describe the display of too strong emotional reactions. In a similar way, the term dissociation was sometimes used without clear criteria and without exploring trauma.

Behavior and symptoms that might indicate trauma were sometimes not understood as such by the staff. Examples are attacks of changed consciousness, violent nightmares, impulsivity, dramatic mood and memory fluctuations, headache and other pain, threatening voices and visions, and depersonalization experiences. Unfamiliar tactile sensations, visions, and voices were generally categorized as psychosis.

Unfamiliar Forms of Expression

The clinicians sometimes reported problems in understanding unfamiliar expressions. In the following example, which may exemplify a cultural idiom of distress, an unfamiliar expression was categorized as noncommunication: "The assessment is that the patient suffers intensely, which she describes as ‘the whole body is burning and aching.' Cannot verbalize; only express herself in this way."

When one woman stated her feeling of having "blood on my hands," this was interpreted as "hysteria." No link was made in this record to the fact that the patient on earlier occasions had spoken about her guilt feelings for the family's asylum situation, which had now abruptly deteriorated.

Unfamiliar Life Circumstances

The complexity of the patients' social situation seemed difficult to grasp for the clinicians. One patient described increasing worries about the asylum process, about her child being in foster care, and about her sick father in the home country. The clinicians interpreted her self-destructiveness as acting out.

In the case of a woman whose asylum process approached deportation, the staff, wishing to help her prepare for this, insisted on contacting the family in the home country. When the woman resisted, explaining that the family had disowned her, this was discussed as a possible "depressive delusion." In the end, the woman accepted a telephone meeting with the father, who explicitly declared that she could not return, since "everyone who gets in touch with her will be socially stigmatized." After the meeting the woman said that she had lost confidence in the clinicians who had subjected her to this humiliation.

Elusive Trauma

This theme refers to the documentation of how trauma was communicated and taken into account in the assessment and treatment processes.

Even when trauma was noted in the medical records it was rarely explored further. Trauma narratives were developed almost exclusively in contacts at outpatient units, especially if the patient had a long-term continuous contact with one clinician. The three patients whose traumatic experiences were extensively addressed in the medical records all had higher education, a political background, and some understanding of the cultural context of their experiences.

One of them gave repeated descriptions of the effects of trauma: not sleeping at all, having constant flashbacks of the rapes, difficulties breathing, popping of the ears, hearing problems, strong heat sensation in the body, constant fear and death thoughts. Another woman was reported to say that, "it feels as if the person who raped her is with her all the time,"

When women who had disclosed traumatic experiences in outpatient contacts were admitted to hospital, the traumatic experiences disappeared from view, overshadowed by their dramatic symptoms and social circumstances, or because the woman did not want to talk about them. There were few examples of the hospital inviting the outpatient clinician to clinical conferences.

Even when the patient received a diagnosis of PTSD, the criteria of this diagnosis seemed not to have been systematically evaluated. The assessment of suicide risk tended to focus on suicide intent, degree of depression, hopelessness, and the attitude to care. The patients themselves, however, often described trauma-related, impulsive suicidal behavior.

In general, the treatment for trauma-related symptoms was medication with antidepressants, hypnotics, and sometimes neuroleptics. Structured activities and physiotherapy were also sometimes recommended. Psychotherapy was generally not regarded as an option owing to the limited access to care in general for asylum seekers.

Meanings of Suicide and Communication of Meaning

This theme refers to the documentation of the meanings patients attached to suicide and suicide attempts.

The records often included an evaluation of a suicidal act based on a suicide risk assessment scale, but more rarely documented the patient's own thoughts about not wanting to live. We noted the following typical comments on the meaning of having attempted suicide: "There is no future," "there is no other way out" were expressions about suicide as a way out of an unsolvable dilemma that was usually linked to the asylum situation. One clinician reported: "She says that she has come to Sweden to start living, and if that is not allowed, she just wants to die." The task of protecting one's children from a dismal future was another unsolvable dilemma for many of the women.

The patients who were parents usually regarded their children as a reason to stay alive. Later in the process, however, some said that the child would be better off without them, sometimes with the addition that the child's chances of asylum might be strengthened if it lost its mother. Three women were reported to consider extended suicide.

Another meaning of suicide was to relieve the woman of immediate pain, to "get rid of the poison inside," as one woman expressed it. This pain tended to be linked to intrusive memories of trauma.

Several women described suicide as a way of being reunited with dead relatives. "We [referring to herself and her daughter] will go to granny, we are not returning to… [home country, omitted]," one woman was reported to say. Staff rarely engaged in dialogue on these or other existential – or religious – issues with the patients, even though practical support for religious activities was given.

Challenges of Combining Perspectives

This theme refers to the documented challenges for the clinicians to embrace multiple perspectives on the patients' experiences and clinical conditions.

When the patients talked about the causes of their mental ill-health, they most often gave social and existential explanations. The clinicians generally assumed biomedical explanations. Social stressors, including the asylum process, were not seen as causal factors. The clinicians repeatedly asked themselves: Is this merely a crisis reaction or a diagnosis proper? In the majority of cases they assessed it as either–or. "The patient's ill-health is assessed to reflect on her life situation at large, and impossible to treat medically," one clinician concluded. Only in very few cases were both explanatory aspects recorded.

Diagnoses, even of the same individual, were frequently changed in the medical records. These changes often seemed related to the patient's social realities. For example, when the asylum situation became more pressing, the diagnosis was sometimes changed from depression or PTSD to adjustment disorder or crisis reaction. The most common diagnoses were depression, PTSD, adjustment disorder, psychosis, and crisis reaction.

Even when social and existential background factors to the patient's condition were prominent, the typical treatment offered was medication and other medical interventions. In some cases, when the clinician was uncertain about crisis reaction or depression, suicide risk seemed to be downplayed, which indicates that suicide was seen as linked to mental illness rather than to social factors.

Shared Powerlessness

This theme refers to repeatedly documented experiences of powerlessness in patients and staff members alike.

One woman was reported to express that, "the more time that passes, the less able she is to control her life." When patients experienced setbacks in the asylum process, their powerlessness was frequently expressed in self-destructive behavior. Confronted with the prolonged asylum process, the clinicians too seemed to feel powerless. When the patient asked for support with her asylum case, the clinician sometimes told her that, "psychiatry has no responsibility." In some cases, the patient was discharged from the hospital when the pressure of the unresolved asylum situation was rising, even shortly after a deterioration of her mental state. "All has been done," was the comment in one record.

Another record stated:

She tries to prove that her situation in the home country is tragic. Pleads for help to be able to stay in Sweden. We inform the patient that we cannot influence the Migration Agency's decision. All we can do is to assign extra staff to guard her.

Other situations that seemed to provoke feelings of powerlessness in the staff were when mothers rejected their children or considered extended suicide, and when patients starved themselves or refused medication.

Powerlessness appeared to be reflected in the clinicians distancing themselves from the patients by the use of value judgments to describe their behavior. Thus, the patient could be described as "demonstrative," "regressive," "unwilling to collaborate" or as not assuming "adult responsibility." A very common expression, which sometimes also came across as a value judgment rather than an aspect of a diagnostic assessment, was that the patient "acted out."

The treatment measures, too, seemed to testify to the clinicians' powerlessness. Electroconvulsive therapy was sometimes described as a last resort, even though the patients did not, according to the records, meet the diagnostic criteria for this intervention. Analogously, "activation," sometimes involving occupational therapy, was repeatedly invoked as a measure against self-starvation.

Discussion

The communication between the asylum-seeking women and the clinicians was profoundly affected by the experience of unbearable realities of the women. There was a sense of shared powerlessness. For the clinicians, powerlessness was related to difficulties in decoding languages of distress, and to the lack of attention to trauma issues, and to the subjective meanings of the suicide attempts. There were difficulties combining the patients' and the clinicians' perspectives.

Lack of Knowledge

Some difficulties seemed to reflect a lack of knowledge in psychiatry at the time of the study period. The medical records indicated a lack of knowledge about trauma-related disturbances such as memory problems and dissociation. Cultural perspectives were also underdeveloped in the diagnostic manuals of the time (American Psychiatric Association, 2000). Currently the use of a Cultural Formulation Interview, exploring the patient's definition of the problem, the patient's perceptions of cause, context, and support, of self-coping and past help-seeking, and of cultural factors affecting current help seeking, is recommended for the assessment of all patients in the DSM-5 (American Psychiatric Association, 2013).

One Bengali woman's state of mind was invariably described as "fear" (a word the interpreter used), but what was meant by fear was never clarified. There are different words in Bengali for fear, related to the intensity as well as the cause or object of the fear reaction. For instance, there is a special word that refers to the fear of losing one's honor and another word for the fear of emptiness, which can mean death but also meaninglessness. There are also fears of bodily phenomena and fears related to spiritual beliefs (H. Bose Hakansson, personal communication, September 19, 2016).

Need of Patient-Centered Focus

The clinicians in this study struggled to understand their patients, but often assessed the patients' reactions against measures of normality that seemed to fit the clinician's earlier experiences. Our findings attest to those of Kirmayer (2003), who wrote that differences in experiential worlds and in knowledge of social contexts may lead to failures of imagination and to refugees facing misunderstanding and incredulity in the psychiatric system.

The subjective meanings of the concerns presented by the patients were rarely explored in the records and not seen as central to the treatment. However, in cases in which the records did convey experienced meanings of suicide, trauma-related impulsive symptoms often came to the fore as a risk factor for suicidal acts. When some clinicians assessed impulse-driven suicide attempts as less serious, this appeared to follow the logic of instruments measuring suicidal intent, for instance, the Suicide Intent Scale (Freedenthal, 2008). A more open exploration of the meaning of suicide for the patients might have enriched the assessment.

The records illustrated the difficulty for patients to disclose and clinicians to explore trauma. Continuity and a therapeutic alliance were necessary for trauma to be genuinely taken into account.

Combining Perspectives

The difficulty of combining contextual and existential perspectives with medical, psychiatric ones was evident in diagnostic processes as well as in treatment plans.

The theory of social causation of mental ill-health and, indeed, of suicide is not new (Durkheim & Simpson, 1897/1952). Concerning the increased risk of suicide in some female migrant groups, cultural conflicts, autonomy issues, and domestic violence have been advanced as explanations (Bhugra et al., 2011; van Bergen, van Balkom, Smit, & Saharso, 2011). However, biomedical models predominate in clinical thinking on suicidality, supported by findings that 90% of all suicides are linked to mental disorder (Mann, 2002). Studies have shown that these figures vary between different countries (Phillips et al., 2002; Radhakrishnan & Andrade, 2012). In our material, the patients often presented a narrative about a desperate life situation, whereas the clinicians' perspective remained dominated by an illness model, which seemed not to support viewing the patient narrative as meaningful. The wish to die expressed by some women was treated as a symptom and not as a personal disclosure.

Traditionally the prospect of being granted asylum in Sweden on grounds of mental ill-health was slim, and in 2006 it was further restricted when the relevant legal statute was revised from offering "asylum on humanitarian grounds" to doing so in "exceptionally distressing circumstances" (). Underscoring that the mental state of the patient had no importance whatsoever for the asylum decision, some clinicians seemed to exaggerate their own powerlessness, and statements like "psychiatry has no responsibility" sometimes appeared to reflect frustration and a wish to maintain distance to the patient's suffering. When descriptions of the patient's reactions were presented out of context in the records, this also seemed to reflect a distancing and sometimes moralizing attitude on the part of the staff.

Consequences for Treatment

Only rarely did the recorded dialogues address what might help the asylum-seeking women handle their unbearable realities. Few systematic psychological interventions were offered to the patients, even during long periods of hospitalization. The fact that so few of the patients were offered psychotherapy can be explained by the general lack of such resources or by a too-restrictive interpretation of asylum seekers' legal access to care. Further, a disinclination on the part of clinicians to initiate therapy in an unstable situation may be misguided. Recent studies have shown that asylum seekers can profit from psychotherapy in spite of their unstable life situation (Başoğlu, Ekblad, Bäärnhielm & Livanou, 2004; Ter Heide, Mooren & Kleber, 2016).

Efforts were made to counteract powerlessness through psychosocial work involving the patients' social networks, including public agencies. The social worker of the ward established collaboration with the social services in all cases of women with children.

Strengths and Limitations

A strength of the study was the systematic procedure of the data analysis, with successive discussions in the research group in each stage of the analysis. We reflected on our own impressions and preliminary understanding of a given experience or interaction and refined the themes, reaching consensus. We used our own experience as two psychiatrists and one psychologist and psychoanalyst, all with extensive clinical experience. The risk, however, was that we would be too accustomed to the perspective of the clinicians to be able to analyze the communication without bias. We dealt with this risk by self-reflection and critical discussions in the group of authors.

Even though the medical records were a rich source of information, they reflected the patient's experience as this was perceived by the clinician. The material also provided indirect information about the clinician: his or her knowledge and attitudes and the impact of the organization. Absent information in the records always raises the question of whether there was nothing to report or whether it was not explored or undocumented.

Conclusion

The findings of this study suggest that clinicians who treat asylum-seeking women who have attempted suicide need to consider the patient's social and cultural context including expectations and roles related to gender. Greater attention to trauma and suicidal pathways in traumatized patients is called for.

The study raises the question of how the integration of context and subjective experiences in psychiatric practice may equip clinicians for the challenges of working with asylum seekers in suicidal crises. This question merits further studies and service development. Approaches that help clinicians to explore the patient's perspective, like the Cultural Formulation Interview (American Psychiatric Association, 2013), could be useful. Prospective studies of both the asylum seekers' and the clinicians' perspectives could contribute to improving mental health care for asylum seekers.