To be granted asylum under the 1951 United Nations Convention Relating to the Status of Refugees, the asylum applicant has to show a ‘well-founded fear of being persecuted in his or her country of origin for reasons of race, religion, nationality, membership of a particular social group, or political opinion’ (United Nations High Commissioner for Refugees, 1992). Since there is often little documentary evidence about the asylum seeker, credibility of the individual is key. Late disclosure, or description of incidents in later interviews of which no mention was made in the first, is commonly cited as a reason to doubt an asylum seeker's credibility (see Asylum Aid, 1999). It is understandable that the addition of new evidence could be seen as evidence against the claimant's honesty. However, this assumption may fail to take into account other reasons for not disclosing at the outset. To date, there has been no empirical study on what affects asylum seekers' disclosure during legal interviews.
Many refugees who come to the UK have experienced or witnessed torture and organised violence (Reference Burnett and PeelBurnett & Peel, 2001). Disclosure is specifically an issue with torture survivors owing to their difficulties of trust in other people (particularly those in authority) and their avoidance of painful memories (Medical Foundation for the Care of Victims of Torture, 2002). A meta-analysis revealed increased prevalence rates of post-traumatic stress disorder (PTSD) in refugees resettled in Western countries (Reference Fazel, Wheeler and DaneshFazel et al, 2005). Symptoms of PTSD may be activated during the Home Office interview as a result of being reminded of the traumatic event, which in turn might reduce a person's ability to give a coherent account and might lead to non-disclosure.
There is also evidence that different trauma types are associated with different PTSD patterns. Two studies found a significant relationship between sexual torture and the avoidance criteria of PTSD (Ramsey et al, 1993; Reference Van Velsen, Gorst-Unsworth and TurnerVan Velsen et al, 1996). Van Velsen et al (Reference Van Velsen, Gorst-Unsworth and Turner1996) speculated that the intimate nature of the sexual attack and associated negative emotions, such as feelings of humiliation and shame, are likely to be critical elements leading to subsequent avoidance behaviour. However, this has not been specifically tested.
Refugees and asylum seekers often come from cultures with different attitudes towards sexuality. Sexual violence and rape are often taboo subjects and can bring about feelings of shame. Women who have been subjected to sexual assault may be shunned by their community and family if they admit to this and therefore may not disclose it in their asylum interview (United Nations, 1997; Reference BurnettBurnett, 1999). Men also tend to underreport experiences of sexual violence (Reference Peel, Mahtani and HinshelwoodPeel et al, 2000). Feelings of shame have been mentioned in the literature as a factor affecting disclosure (Reference Hill, Thompson and CogarHill et al, 1993) and there have been several empirical studies demonstrating the relationship between shame and disclosure (Reference Swan and AndrewsSwan & Andrews, 2003; Reference Hook and AndrewsHook & Andrews, 2005). There is also increasing evidence that shame may be linked to the course or onset of PTSD (Reference Andrews, Brewin and RoseAndrews et al, 2000; Reference Leskela, Dieperink and ThurasLeskela et al, 2002).
The study of different trauma types by Van Velsen et al (Reference Van Velsen, Gorst-Unsworth and Turner1996) suggested including a measure of dissociative phenomena in future research, as dissociation might be closely related to PTSD avoidance symptoms. Indeed, dissociative experiences are commonly reported by individuals with a diagnosis of PTSD (Reference Ozer, Best and LipseyOzer et al, 2003). Carlson & Rosser-Hogan (Reference Carlson and Rosser-Hogan1991) found high levels of association between traumatic experiences and the severity of both traumatic stress and dissociative reactions in a group of 50 Cambodian refugees. However, dissociative responses not only occur as an aftermath of a traumatic event, but can also be experienced at the time of the trauma (peritraumatically; Reference Weiss, Marmar and MetzlerWeiss et al, 1995). Dissociative reactions might be activated during an anxiety-provoking event, such as the Home Office interview, which might affect disclosure.
The first aim of our study was to investigate the impact of sexual violence on refugees' and asylum seekers' reported post-traumatic stress symptoms, shame reactions, dissociative experiences and difficulties in disclosure during Home Office interviews. The second aim of the study was to explore more systematically the factors involved in refugees' and asylum seekers' disclosure during Home Office interviews by means of a qualitative semi-structured interview.
METHOD
Sample and procedure
Refugees and asylum seekers with a history of pre-migration trauma were included in the study. Twenty-seven participants in total were recruited from a central London traumatic stress clinic (n=17) and two London-based community services (n=10). They were invited to take part in a research study about refugees' and asylum seekers' experiences of legal interviews; demographic data are reported in Table 1. The participants, who had arrived in the UK between 1995 and 2003, originated from 14 countries in Europe, Africa, the Middle East and Latin America. Written informed consent was obtained.
Comparison of groups by measures
| Sexual violence (n=15) | Non-sexual violence (n=12) | t(21–25) | |
|---|---|---|---|
| Gender, n | |||
| Male | 4 | 7 | |
| Female | 11 | 5 | |
| Asylum status, n | |||
| ILR | 6 | 8 | |
| ELR | 2 | 1 | |
| Under appeal | 7 | 3 | |
| Asylum decision following Home Office interview, n | |||
| Yes | 6 | 7 | |
| No | 9 | 5 | |
| Psychological treatment, n | |||
| Yes | 10 | 5 | |
| No | 5 | 7 | |
| Age, years: mean (s.d.) | 37 (12.0) | 45.4 (12.2) | –1.80 |
| Time lag in months: mean (s.d.) | 40.5 (25.3) | 51.5 (26.9) | –1.06 |
| Scores: mean (s.d.) | |||
| PSS–I | |||
| Overall severity | 37.7 (10.7) | 27.1 (11.6) | 2.46* |
| Re-experiencing | 8.9 (3.0) | 9.0 (4.1) | –0.05 |
| Avoidance | 16.0 (4.4) | 7.8 (5.4) | 4.37*** |
| Hyperarousal | 12.7 (4.5) | 10.3 (5.0) | 1.31 |
| HSCL depression | 43.5 (11.4) | 36.3 (10.7) | 1.68 |
| ESS | 65.6 (19.6) | 42.2 (9.2) | 4.10** |
| PDEQ–SRV | 31.9 (10.1) | 20.0 (13.3) | 2.84* |
| Difficulty in disclosure | 3.5 (0.9) | 1.7 (1.0) | 4.91*** |
ELR, exceptional leave to remain; ESS, Experience of Shame Scale; HSCL, Hopkins Symptom Checklist; ILR, indefinite leave to remain; PDEQ–SRV, Peritraumatic Dissociative Experiences Questionnaire – Self-Report Version; PSS–I, PTSD Symptom Scale – Interview
At the time of testing, 15 of the 27 study participants were receiving psychological input at a specialist tier 3 London traumatic stress clinic. Nine of them were receiving long-term weekly individual psychological treatment, and 6 had just completed a 3-month weekly psychoeducation group. This group was run for people who were on the waiting list for individual psychology. The remainder (n=12) had not received psychological input since coming to the UK.
Participants were divided into two groups. The first group consisted of participants with a history of sexual violence. Following the study by Van Velsen et al (Reference Van Velsen, Gorst-Unsworth and Turner1996), sexual violence was defined as rape (of men or women) or other tortures directed to the genital area. The second group consisted of participants with a history of non-sexual violence. This was broadly defined as having experienced or witnessed some form of psychological and/or physical maltreatment including torture. Overall, 15 participants experienced some form of sexual violence, including rape (n=12) and sexual torture (n=3). Twelve participants experienced or witnessed some other form of violence, including torture (n=6), being shot (n=2), beatings (n=2) and witnessing killing of family members (n=2). This information was obtained, with consent, from the person's clinician or caseworker, or from medical notes. All participants had had a screening interview shortly after their arrival in the UK, followed by one or more main Home Office interviews; 24 participants had had one main Home Office interview and 3 participants had had two main Home Office interviews.
Research interviews took place over a 6-month period from November 2004 to May 2005. Participants were interviewed on one occasion about their main Home Office interview. People who had attended two main Home Office interviews were questioned about their first one. Interpreters were used when requested by participants. Seven participants were interviewed with the assistance of an interpreter who was officially accredited. To avoid translation issues, all measures were presented orally during the interview.
Measures
PTSD Symptom Scale-Interview
The PTSD Symptom Scale-Interview (PSS-I; Reference Foa, Riggs and DancuFoa et al, 1993) was used to assess current PTSD symptoms according to DSM-IV criteria (American Psychiatric Association, 1994). This is a semi-structured interview consisting of 17 items; answers are rated from 0 (not at all) to 3 (five or more times per week/very much). Total severity scores are based on the sums of the raw items.
Hopkins Symptom Checklist
The Hopkins Symptom Checklist-25 (HSCL-25; Reference Derogatis, Lipman and RickelsDerogatis et al, 1974) was included since depression has been found to be highly comorbid with PTSD (Reference Blanchard, Buckley and HicklingBlanchard et al, 1998). The HSCL-25 was chosen for its cross-cultural robustness (Reference Kinzie and MansonKinzie & Manson, 1987). Participants completed part 2 of the scale, which has 15 depression items rated on a four-point scale, ranging from 1 (not at all) to 4 (extremely). The mean of the 15 depression items has been shown to correlate with major depression as defined by the DSM-IV (American Psychiatric Association, 1994).
Experience of Shame Scale
The Experience of Shame Scale (ESS; Reference Andrews, Qian and ValentineAndrews et al, 2002) is a 25-item scale assessing three different domains of shame: characterological, behavioural and bodily shame. Within each of these domains there are items reflecting the experiential (feeling shame), cognitive (concern over others' opinions) and behavioural (concealment or avoidance) components of shame. Participants rate each item according to how they have felt in the past year, on a four-point scale ranging from 1 (not at all) to 4 (very much).
Peritraumatic Dissociative Experiences Questionnaire
The Peritraumatic Dissociative Experiences Questionnaire-Self-Report Version (PDEQ-SRV; Reference Marmar, Weiss, Metzler, Wilson and KeaneMarmar et al, 1997) consists of ten items measuring retrospectively acute dissociative reactions during a specific event. Items are rated on a five-point scale, ranging from 1 (not at all true) to 5 (extremely true). Participants were instructed to complete the items based on their experiences and reactions during the Home Office interview and immediately afterwards.
Difficulty in disclosure
Participants were asked to rate on a four-point scale, ranging from 1 (not at all) to 4 (extremely), how difficult they found it to disclose personal information during the Home Office interview.
Semi-structured interview
A semi-structured interview was used to collect qualitative data regarding people's disclosure during Home Office interviews. Because of ethical constraints we did not set out to investigate whether sexual victimisation was disclosed or not during the Home Office interviews. Interviews were taped and transcribed. Four participants did not want their interview recorded and in these cases process notes were taken instead. Participants were asked a number of general questions relating to the disclosure of their index trauma.
-
(a) When was the first time you talked about what happened to you in (your home country)? After the event? After your arrival in the UK?
-
(b) Who did you talk to?
-
(c) Was there anything you initially did not tell this person?
Other questions specifically concerned disclosure behaviour during the Home Office interview.
-
(d) To what extent did you feel you could open up and talk openly about what happened?
-
(e) Are there any things you have not yet told the Home Office about? If yes, could you tell me what some of the reasons might be that you have found it difficult to do that?
Finally, a question was included to assess whether participants could identify any aspects relating to their cultural background that had affected disclosure during their Home Office interview, because research has shown that issues such as sexual violence are not readily disclosed to others owing to feelings of shame, social stigma and the risk of being shunned by family members and the community.
-
(f) Are there things you have not talked about because in your culture it is considered wrong?
Other questions assessed participants' experiences of the Home Office interview, particularly addressing interpersonal and situation- and context-specific factors, as well as other issues and recommendations. These data will be reported separately.
Demographic and clinical factors
Demographic data were collected for all participants, including age, gender, nationality, current asylum status, dates of arrival in the UK, number and dates of Home Office interviews, decision on asylum claim following Home Office interview, time elapsed between Home Office interview and research interview (in months), and receipt of psychological treatments since arrival in the UK.
Statistical analysis
Several variables had skewed distributions and required transformation. Following Tabachnick & Fidell (Reference Tabachnick and Fidell2001), analyses using untransformed data are reported, as transformation did not affect the results. Differences between sexual and non-sexual violence groups on demographic factors and measures of PTSD, depression, shame, dissociation, and difficulty in disclosure were investigated using independent t-tests. Analysis of covariance was used to control separately for the effects of relevant variables on group differences in difficulty of disclosure. Correlations between age, time lag between Home Office and research interviews, PTSD, depression, shame, dissociation and difficulty in disclosure were examined using Spearman's rho. Partial correlations were used to determine whether that the associations between dissociation and shame and dissociation and disclosure were still significant after total PTSD symptoms were controlled for. Independent t-test was used to measure the relationship between difficulty in disclosure and decision on asylum claim, as well as receipt of psychological treatments. Statistical analyses used the Statistical Package for the Social Sciences version 11.5 for Windows. A two-tailed α level of P=0.05 was used to determine statistical significance.
The qualitative data were analysed using a thematic analysis approach, which focuses on identifiable themes and patterns of personal experiences (Reference AronsonAronson, 1994). Following recommendations by Elliot et al (Reference Elliot, Fischer and Rennie1999), credibility checks were provided in several ways. To provide checks on reliability, a second marker audited the data from each question, looking at the themes created. Any differences in opinion were discussed and rectified. Furthermore, the findings were triangulated by comparing the outcome of the qualitative data with the results of the quantitative data and drawing parallels between the two (see Discussion). The validity of the conclusions drawn from the interview data is enhanced in several ways: first, we present direct quotes from the interviews to demonstrate to the reader the relationship between themes and the source data; second, to indicate how representative the themes were of the sample as a whole, the proportion of participants for each theme is outlined; and third, the analysis includes a negative case analysis, which means reporting on minority as well as majority responses.
RESULTS
Quantitative findings
No significant group difference existed for age, time lag in months between participants' main Home Office interview and research interview, PTSD re-experiencing symptoms, PTSD arousal symptoms or depression (Table 1). Those with a history of sexual violence reported greater overall PTSD severity and avoidance symptoms, as well as greater feelings of shame (Table 1). This group also described more dissociation symptoms and greater difficulty in disclosure of personal information during their Home Office interview.
There was no association between self-disclosure and decision on asylum claim following the Home Office interview (t(25)=-0.78; P>0.05) or between self-disclosure behaviour and receipt of psychological treatments (t(25)=0.89; P>0.05).
Those with higher levels of shame also had higher PTSD scores and showed increased avoidance and arousal symptoms (Table 2). No significant relationship existed between total shame scores and PTSD re-experiencing symptoms. Respondents with increased dissociation scores had higher levels of shame and showed greater PTSD avoidance symptoms. Greater difficulty in disclosure was positively associated with higher levels of PTSD total scores, PTSD avoidance symptoms, shame, depression and dissociation, but not with time lag.
Intercorrelations among measures using Spearman's rho (n=27).
| Measures | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| 1. Age | ||||||||||
| 2. PSS–1 Overall severity | –0.44 * | |||||||||
| 3. PSS–1 Re-experiencing | –0.13 | 0.66 *** | ||||||||
| 4. PSS–1 Avoidance | –0.47 * | 0.82 *** | 0.26 | |||||||
| 5. PSS–1 Arousal | –0.28 | 0.84 *** | 0.68 *** | 0.45 * | ||||||
| 6. ESS | –0.24 | 0.75 *** | 0.26 | 0.79 *** | 0.52 ** | |||||
| 7. HSCL depression | –0.35 | 0.80 *** | 0.65 *** | 0.59 ** | 0.74 *** | 0.58 ** | ||||
| 8. PDEQ–SRV | –0.23 | 0.42 * | 0.06 | 0.44 * | 0.15 | 0.61 ** | 0.25 | |||
| 9. Difficulty in disclosure | –0.40 * | 0.55 ** | 0.07 | 0.63 *** | 0.37 | 0.69 *** | 0.51 ** | 0.79 *** | ||
| 10. Time lag | 0.19 | –0.21 | –0.24 | –0.10 | –0.19 | –0.05 | –0.32 | –0.18 | –0.32 |
ESS, Experience of Shame Scale; HSCL, Hopkins Symptom Checklist; PDEQ–SRV, Peritraumatic Dissociative Experiences Questionnaire – Self-Report Version; PSS–1, PTSD Symptom Scale – Interview
* P<0.05
** P<0.01
*** P<0.001 (all tests were two-tailed)
A series of analyses of covariance were carried out to explore whether the groups still differed on difficulty in disclosure when controlling for PTSD overall severity, PTSD avoidance symptoms, shame and dissociation. The results remained unchanged, showing that there still was a significant difference between groups on self-disclosure behaviour after controlling for the effects of PTSD overall severity (F (1,24)=14.75, P<0.01), PTSD avoidance symptoms (F (1,24)=7.13, P<0.05), shame (F (1,24)=7.70, P<0.05) and dissociation (F (1,24)=13.13, P<0.01). This indicates that none of these factors on its own was responsible for the effect. Since dissociative experiences are a diagnostic feature of PTSD, partial correlations were carried out showing that the associations between dissociation and shame (r=0.78, P<0.001) and dissociation and disclosure (r=0.60, P<0.01) were still significant after total PTSD symptoms were controlled for.
Qualitative findings
General disclosure behaviour
Twenty participants out of 27 reported that the first time they talked about the traumatic event was after their arrival in the UK; the majority of those talked to Home Office officials (n=13), the rest talked to family members (n=3), healthcare professionals (n=2) or their solicitor (n=2). Out of the 14 people who disclosed to others than the Home Office, 10 reported that they initially did not tell the person everything. Reasons cited included the impact of past traumatic events, such as feelings of confusion and shock (n=3), a need to build up trust and confidence before being able to talk about sexual issues (n=3); feeling scared that details might be passed on to their government or that they would not be believed (n=3), and not wanting to burden other family members (n=1).
Disclosure behaviour during Home Office interview
Three different themes emerged from the participants' answers: no reported problem in opening up; finding it too difficult to disclose; and wanting to disclose, but not being given the chance to do so. Seven people reported no difficulties with opening up and disclosing personal details in their Home Office interview. Twelve people reported difficulties in disclosing personal details during the Home Office interview; 10 of them had a history of sexual violence. Reasons cited were feeling too traumatised, afraid and ashamed to talk about the past (n=10), which resulted in them not being able to tell the Home Office interviewer what had happened to them or to answer questions.
‘It was the first time in my life that I had to talk about what happened to me. I only told the interviewer about 10%, I could not talk, it was too difficult. I felt so traumatised and ashamed.' (P2)
Further reasons cited were intrusive experiences, such as intrusive memories and flashbacks, which affected their ability to focus on the interview and give a coherent account (n=2):
‘When I talked about the past, what happened to me, the memories came, flashbacks. And then I found it difficult to remember anything that happened in my country. I was crying, I was shocked. It was hard to explain what happened to me.' (P1)
Others reported dissociative experiences that made it difficult for them to focus on the interview, and affected their ability to disclose:
‘I tried to talk, but my mind kept wandering off and I kept thinking about the trauma and my family that I lost. Everything seemed unreal to me, I felt like I was dreaming. I found it hard to focus on the interview and answer questions.' (P6)
Ten people reported that they wanted to tell the Home Office what had happened to them, but that they were not given the opportunity to do so; the interviewer was apparently more interested in factual details about their home country and how they got to the UK than what had happened to them or their families:
‘I wanted to explain properly, but they just stopped me. They ask you to make it short and give yes or no answers. You don't get a chance to say much or explain to them. Therefore I did not go into much detail. But that affected me later [at the court] when I was asked why I did not tell them in the [Home Office] interview.' (P15)
Five of the people who wanted to disclose also reported that they were asked similar questions repeatedly, which increased their stress levels and affected on their ability to disclose:
‘When he asked me questions and I answered them he started cross-examining me. The more I said the more questions he asked me. It felt like he was trying to trick me. I felt nervous and stressed, which made it harder to talk for me.' (P16)
Fifteen people reported that there are still things they have not told the Home Office about; 10 were men and women with a history of sexual torture, and most of them reported feelings of shame as a reason for non-disclosure (n=7):
‘I wanted to keep things from my past private. I was scared that they would look at me badly and make me feel ashamed. I could not tell everything at the interview, but later on I was able to tell the court. They were nice at the court and made me feel more relaxed.' (P21)
Other reasons included forgetting some details, which they were not able to mention subsequently in later interviews for fear it would affect their credibility (n=2); being unsure whether they could disclose details they were not directly asked about (n=3); and not being given the opportunity and the time to talk openly about their past traumatic experiences (n=2).
Cultural factors affecting disclosure
Eight participants reported that there were things they have not talked about because in their culture it is considered wrong; all of them were men and women with a history of sexual violence. Most of them stated that in their culture sexual issues are not talked about, especially rape:
‘At home you are not allowed to talk to other men you are not related to, you are not allowed to look any men in the eyes. So how could I have looked him [male Home Office official] in the eyes and told him what happened to me - it's a different culture.' (P11)
Two individuals specifically mentioned feelings of shame associated with rape, and that shame had prevented them from talking about the rape in the interview:
‘There is a lot of shame associated with what I experienced. Shame in my culture prevents me from talking about this.' (P17)
Direct disclosure of sexual victimisation
Although data on disclosure of sexual victimisation were not specifically collected, further analysis of the transcripts revealed that of the 15 people with a history of sexual violence, 5 reported that they had disclosed sexual victimisation, including rape, during their Home Office interview, and 6 did not disclose it. It is unclear whether the remaining 4 specifically disclosed sexual victimisation. Interestingly, everybody who disclosed a history of sexual violence reported being prevented from talking about it further in the interview by the Home Office official.
DISCUSSION
This study refined and extended previous findings by Van Velsen et al (Reference Van Velsen, Gorst-Unsworth and Turner1996) by demonstrating that there is a significant association between shame and PTSD avoidance symptoms, which suggests that shame might act as a mediator between a history of sexual violence and PTSD avoidance symptoms. Shame was also significantly associated with overall PTSD severity, which provides further evidence that shame might be linked to the course and onset of PTSD (Reference Andrews, Brewin and RoseAndrews et al, 2000; Reference Leskela, Dieperink and ThurasLeskela et al, 2002). The significant relationship between dissociation and PTSD avoidance symptoms confirms speculations by Van Velsen et al (Reference Van Velsen, Gorst-Unsworth and Turner1996). The results are also in line with research showing that dissociative experiences are commonly reported by individuals with a diagnosis of PTSD (Reference Ozer, Best and LipseyOzer et al, 2003). Furthermore, our analysis revealed that those who experienced higher levels of dissociative experiences during the Home Office interviews were those who had higher levels of shame.
Data from the qualitative interviews provide further evidence for the above findings. Perhaps one of the most striking findings was that 20 participants talked for the first time about their pre-migration trauma only after entering the UK, and of those, 13 talked to Home Office officials. These findings underscore the degree of avoidance associated with the experience of trauma and are likely to be very relevant to the large numbers of refugees coming to the UK who have experienced or witnessed torture and organised violence (Reference Burnett and PeelBurnett & Peel, 2001).
Many participants reported difficulties with disclosing personal details in their Home Office interview, and reasons frequently cited for this were negative emotions such as feeling too traumatised by past experiences or feelings of shame. Shame was especially salient for people with a history of sexual violence. Many of those reported that in their culture sexual issues are not discussed with others, and that this prevented them from disclosing sexual issues during their Home Office interview. This supports previous findings that shame is associated with difficulty in disclosure (Reference Swan and AndrewsSwan & Andrews, 2003; Reference Hook and AndrewsHook & Andrews, 2005) and is consistent with Hill et al (Reference Hill, Thompson and Cogar1993) who found that sexual issues often remain too shameful to discuss, even in therapy.
Participants also reported experiencing psychological symptoms during Home Office interviews, such as dissociative experiences, flashbacks and avoidance behaviours (e.g. avoiding thoughts or feelings associated with the trauma and not being able to remember details), which had an impact on their ability to disclose. This suggests that people's psychological states should be routinely evaluated when assessing their ability to give a coherent personal history in an interview with officials.
Finally, it should be noted that although the difficulties with disclosure seemed to be persistent, many participants did express a willingness to talk to officials about their experiences. However, some described not being given the opportunity to do so or being prevented by the interviewer from discussing their experiences. One explanation could be vicarious traumatisation of the interviewers, which is a common phenomenon in people working with trauma survivors (Reference FigleyFigley, 1995). Indeed, a multidisciplinary analysis of the decision-making process of the Canadian Immigration and Refugee Board showed that coping with vicarious traumatisation and uncontrolled emotional reactions was one of the factors having a negative impact on the board members' ability to evaluate credibility and on the overall conduct of hearings (Reference Rousseau, Crepeau and FoxenRousseau et al, 2002). This needs to be clarified by further research.
In summary, our results indicate that late disclosure or non-disclosure during Home Office interviews does not necessarily imply a lack of honesty on the asylum seeker's part, and highlight that disclosure is complex and influenced by a variety of factors that need to be taken into account when judging asylum seekers' credibility based on the information they disclose. A Home Office interview can be a stressful and anxiety-provoking event, which may provoke reactions that interfere with disclosure.
Limitations of the study
Several methodological aspects of this study warrant consideration. The sample size was small, which ruled out the use of multivariate analyses. Language and cultural barriers presented an obstacle, as they made the collection of accurate data more difficult and might have increased measurement error. There is also the potential for a sampling bias, especially finding a sample that is representative of the general refugee and asylum-seeker population. However, this is an applied study of a real life situation, representing the diverse population of refugees going through asylum interviews in the UK. Van Velsen et al (Reference Van Velsen, Gorst-Unsworth and Turner1996) suggested that sampling biases generally pose a problem in research studies on refugees and asylum seekers, as this population is already exposed to numerous selection biases. Similarly, the lack of a control group restricts our findings. The comparisons are limited because the base rates of PTSD, shame, depression, dissociation and difficulty in disclosure are unknown in this group. It would, of course, be desirable to find a comparison group of refugees and asylum seekers who had not experienced any kind of violence. Whether there are refugees and asylum seekers who fit these criteria depends largely on the definition of violence and the definition of ‘refugee’ itself. None the less, the above issues restrict the generalisability of the findings and the tentative conclusions outlined in this paper should be considered with this in mind.
Another limitation concerns people's accuracy in reporting emotional experiences that occurred several months or even years ago. However, since there is no significant difference between groups in the length of time between Home Office interviews and research interviews, this is unlikely to affect the interpretation of the data significantly. On a similar note, dissociation may in some cases have been experienced after the interview. Finally, the cumulative effect of multiple traumas needs to be considered; the greater difficulty in disclosure in the sexual violence group may be related to the fact that some people with a history of sexual violence also experienced physical trauma.
Implications of our findings
The above findings have implications for the process of granting asylum in the UK. Asylum seekers often come from countries where they experienced or witnessed torture and organised violence, which means that they are in a vulnerable position when entering the UK. Most asylum seekers in our study experienced the immigration process - including the Home Office interviews - as stressful and anxiety-provoking, because many feared deportation. Disclosure is a difficult issue in this group; many need time to process past traumatic events and to establish a sufficient level of trust and confidence to reveal the potentially painful and shaming details of their experiences. This needs to be taken into account by an immigration system that requires asylum seekers to make a claim shortly after arrival. It is therefore of paramount importance that sensitivity is used when processing refugee claims and that immigration officials are aware of the needs of asylum seekers in order to avoid inducing further distress in this already highly traumatised group.
The findings also have implications for current immigration policy. The need for policies that identify asylum seekers who fabricate their stories and that deter immigrants who have left their country for economic reasons seems understandable. However, this study suggests that legitimate asylum seekers may be punished and retraumatised by the enforcement of some of these policies. Furthermore, the immigration system needs to take into account the special needs of victims of sexual violence, particularly since there is a high incidence of shame in this group. Given the significant associations between shame, PTSD avoidance symptoms and difficulty in disclosure, one might speculate that being forced to talk about a traumatic event could potentially activate shame reactions, and that people experiencing more shame are engaging in strategies to avoid this feeling, such as non-disclosure of sensitive personal information. This also highlights the importance of recognising and dealing with asylum seekers' shame in an empathic way. It seems that immigration officials could benefit from supervision and training in how to recognise stress reactions in interviewees, such as PTSD symptoms, shame and dissociative experiences, as well as an awareness of the impact of these on people's psychological health, affective states and ability to disclose.
Acknowledgements
The University College London Central Research Fund is gratefully acknowledged for funding this study. We also thank Dr Stuart Turner for his very helpful comments on this paper. This work was undertaken with the support of Camden and Islington Mental Health and Social Care Trust, who received funding from the NHS Executive.
eLetters
No eLetters have been published for this article.