A mixed-method study of expert psychological evidence submitted for a cohort of asylum seekers undergoing refugee status determination in Australia

lable at ScienceDirect

Social Science & Medicine 98 (2013) 106e115

Contents lists avai

Social Science & Medicine

journal homepage: www.elsevier .com/locate/socscimed

A mixed-method study of expert psychological evidence submitted fora cohort of asylum seekers undergoing refugee status determinationin Australia

Kuowei Tay a, *, Naomi Frommer a, Jill Hunter b, Derrick Silove a, Linda Pearson b,Mehera San Roque b, Ronnit Redman b, Richard A. Bryant c, Vijaya Manicavasagar a,Zachary Steel a

a Psychiatry Research and Teaching Unit, School of Psychiatry, University of New South Wales, NSW, Australiab Faculty of Law, University of New South Wales, NSW, Australiac School of Psychology, University of New South Wales, NSW, Australia

a r t i c l e i n f o

Article history:Available online 18 September 2013

Keywords:RefugeesRefugee status determinationAdministrative decision makingAsylum seekersPosttraumatic stress disorder

* Corresponding author. Centre for Population MenHospital, Cnr, Forbes and Campbell Streets, Liverpool

E-mail addresses: [email protected], tayku94

0277-9536/$ e see front matter � 2013 Elsevier Ltd.http://dx.doi.org/10.1016/j.socscimed.2013.08.029

a b s t r a c t

The levels of exposure to conflict-related trauma and the high rates of mental health impairmentamongst asylum seekers pose specific challenges for refugee decision makers who lack mental healthtraining. We examined the use of psychological evidence amongst asylum decision makers in New SouthWales, Australia, drawing on the archives of a representative cohort of 52 asylum seekers. A mixed-method approach was used to examine key mental health issues presented in psychological reportsaccompanying each asylum application, including key documents submitted for consideration of asylumat the primary and review levels. The findings indicated that the majority of decision makers at bothlevels did not refer to psychological evidence in their decision records. Those who did, particularly in thecontext of negative decisions, challenged the expert findings and rejected the value of such evidence.Asylum seekers exhibiting traumatic stress symptoms such as intrusive thoughts and avoidance, as wellas memory impairment, experienced a lower acceptance rate than those who did not across the primaryand review levels. The findings raise concern that trauma-affected asylum seekers may be consistentlydisadvantaged in the refugee decision-making process and underscore the need to improve the under-standing and use of mental health evidence in the refugee decision-making setting. The study findingshave been used to develop a set of guidelines to assist refugee decision makers, mental health pro-fessionals and legal advisers in improving the quality and use of psychological evidence within therefugee decision-making context.

� 2013 Elsevier Ltd. All rights reserved.

Introduction

In 2012, there were 479,300 asylum applications registeredacross Europe, North America, and Australia, the second highestnumber recorded in the last decade (UNHCR, 2012a, 2012b). Underinternational law, a refugee is defined as a person outside of his orher country of origin andwho is unable to avail him or herself of theprotection of and return to that country, owing to a well-foundedfear of persecution for reasons of “race, religion, nationality,membership of a particular social group or political opinion”. Non-

tal Health Research, Liverpool, NSW 2170, [email protected] (K. Tay).

All rights reserved.

refoulement, a key provision in the 1951 Geneva ConventionRelating to the Status of Refugees, proscribes the removal of refu-gees to their country of origin where their safety could bethreatened.

The Indochinese refugee crisis during the late 1980s led to thewidespread implementation of refugee status determination pro-cedures to assess the claims of asylum seekers. This occurred inresponse to growing concern amongst resettlement nations, that agrowing proportion of asylum seekers were seeking economicopportunities rather than protection from persecution (Robinson,1998). Since this time, formal procedures to assess asylum claimshave been established by the UNHCR and by the vast majority ofcountries involved in the permanent resettlement of refugee pop-ulations (Hamlin, 2012).

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115 107

Within Australia, the site of the current research, refugee pro-tection applications are assessed within a two-tier structure,initially by officials of the Department of Immigration and Citizen-ship (DIAC), which we refer to hereafter as the primary level. If theapplicant receives a negative outcome at the primary level they canhave their protection claim reassessed by the Refugee andMigrationReview Tribunal (RRT/MRT) (we refer to this decision-making stageas the review level) (Hunter, Pearson, San Roque, & Steel, 2013;Kneebone, 1998). At each stage, the asylum seeker is required tomake a written application and may also be required to give oralevidence in support of their claim during hearings or interviews.The decisionmaker will determinewhether the applicant is eligiblefor the grant of a protection visa based on review of the written andoral statements and supplementary evidence, including countryinformation from government and non-government organizations,case law, and forensic information such asmedical, psychological orlinguistic evidence. Decision makers at both primary and reviewlevels face complex and substantial challenges in reaching a deci-sion in that they must evaluate not only any evidence, but also thecredibility of the applicant and the plausibility of their account,whether they are likely to face persecution if returned home, andthe possibilities of internal relocation or alternative options for theperson within the country of origin.

There is a growing recognition that asylum seekers invariablyface difficulties in providing evidence to support their claims ofpersecution since they are often unable to obtain documentaryevidence of persecution upon fleeing the country of origin (Cohen,2001b; Dauvergne & Millbank, 2003). As a consequence, decisionmakers in asylum recipient countries have tended to focus onconsistency, standard disclosure, and demeanor as key criteria forassessing the credibility of an applicant’s claim (Cohen, 2001a,2001b; Kagan, 2003; Kneebone, 2003; Macklin, 2006; Millbank,2009). In response to the reliance on such assessments, mentalhealth professionals have raised particular concern about the ef-fects of traumatic exposure on the testimonies of asylum seekers,relating to memory impairment, presentation, and reporting ofinformation (Herlihy, Gleeson, & Turner, 2010; Herlihy & Turner,2006, 2009; Meffert, Musalo, McNiel, & Binder, 2001).

In particular, there is concern that exposure to refugee-relatedtraumas and the psychiatric sequelae associated with such expo-sure may adversely impact the refugee decision-making process.For the most part, refugee decision makers will lack specializedmental health training and as such may interpret mental healthsymptoms as evidence undermining an applicant’s credibility(Prabhu & Baranoski, 2012). Research undertaken with refugeepopulations supports the general concern that the presence ofposttraumatic stress symptoms is associated with memoryimpairment. For example, Herlihy, Scragg, and Turner (2002) in aUK-based study of refugees subjected to repeated interviewsdocumented a relationship between severity of posttraumaticstress symptomatology and narrative discrepancies in the refugees’accounts. In particular, they found greater discrepancies concerningdetails considered as peripheral rather than central to the coretraumatic narrative (Herlihy et al., 2002). Although there have beensome contrary research findings (Eytan, Laurencon, Durieux-Paillard, & Ortiz, 2008), data from available studies broadly sup-port this observation. For example, Moradi et al. (2008) found thatseverity of posttraumatic stress symptoms is associated withreduced specificity of autobiographical memories in refugees(Moradi et al., 2008). Traumatic events involving sexual violencehave also been found to be associated with limited disclosure ofsensitive information as a result of interpersonal, cultural, andpsychological factors, which may impact on the decision-makingprocess (Bogner, Brewin, & Herlihy, 2010; Bogner, Herlihy, &Brewin, 2007).

Findings from research with other trauma-affected populationgroups have identified a broad array of cognitive and memorydomains that show evidence of impairment. For example, in-dividuals exhibiting posttraumatic stress symptoms have aheightened tendency to report overgeneralized memory that lacksspecificity compared to those who do not display such symptom-atology (McNally, Lasko, Macklin, & Pitman, 1995). Trauma survi-vors also show fragmentation and disorganization in theirtraumatic memories (Foa, Molnar, & Cashman, 1995; King et al.,2000; Southwick, Morgan, Nicolaou, & Charney, 1997), a findingthat also holds for adverse non-traumatic memories in traumasurvivors (Jelinek, Randjbar, Seifert, Kellner, & Moritz, 2009).

There is a large body of evidence from clinical and experimentalstudies that trauma-affected individuals display a bias in recall forimportant or central details of an event at the expense of peripheralor irrelevant materials (Brewin, 2011), although some research hasalso found that memory for traumatic events (e.g. sexual violence)is enhanced, rather than impaired, under conditions of extremestress (Bernsten & Rubin, 2007, McNally, 2003; Porter & Birt, 2001).A meta-analytic review of 27 studies of the effects of stress oneyewitness testimony identified a relationship between levels ofanxiety and impairedmemory recall for both central and peripheraltypes of information, suggesting that stress may disrupt the con-struction of mental representations of complex emotional scenes(Warring, Payne, Schacter, & Kensinger, 2010). These conflictingfindings might reflect methodological differences in assessingstress and approaches to defining memory (Brewin, 2011). Somestudies have investigated the role of dissociation in precipitatingand maintaining PTSD. In particular, the presence of self-reporteddissociation has been found to be a significant predictor of frag-mentation in trauma narratives (Brewin, 2007). Furthermore,engaging in avoidance of trauma-related thoughts and memories, akey feature of PTSD, has been shown to be associated with severityof PTSD symptoms over time (Ehlers, Mayou, & Bryant,1998; Josephet al., 1996).

In concert, these findings raise concern about the possibleimpact of posttraumatic stress and mental health impairment onasylum testimonial evidence and subsequent refugee decisionmaking. Observations frommental health professionals in refugee-forensic mental health settings highlight the potential risks ofrefugee decision makers interpreting mental health issues asundermining asylum seekers’ credibility (Prabhu & Baranoski,2012; Steel, Frommer, & Silove, 2004; Turner & Herlihy, 2009).Prabhu and Baranoski (2012) report that a traumatic history isroutinely narrated with the omission of important details and/orwith a flat affect as a result of the symptoms of numbing andavoidance, which may raise suspicion amongst decision makers.Cultural factors also play a prominent role in many aspects ofnarrating a traumatic event (e.g., cultural differences in temporalorientation; disclosure of sexual violence) and in the psychiatricpresentation of the traumatized asylum seeker. In the refugeedetermination context, decision makers have demonstrated a ten-dency to interpret narrative inconsistencies in asylum seekers’ ac-counts as evidence of lack of credibility (Masinda, 2004; Rousseau,Crepeau, Foxen, & Houle, 2002). Rousseau et al. (2002) in a seminalstudy of adverse refugee status decisions by Canadian decisionmakers identified that cultural, psychological, and systemic issues(e.g., lack of knowledge about psychiatric expertise, cultural mis-understandings) exerted a negative impact on the refugee decision-making process. Masinda (2004) identified an association betweenasylum seekers exhibiting posttraumatic stress symptoms andadverse decision outcomes. Legal professionals within the forensiccontext have also been found to exhibit misunderstandings aboutthe nature of eyewitness testimony (Granhag, Stromwell, &Hartwig, 2005).

Table 1Asylum decisions included in the present analysis.

Primary level Review level

Total number of primarydecisions issued after3 years[ 66 involving70 asylum seekers

Total number of decisions issuedafter 3 years[ 44 involving 46asylum seekers who appealed theirprimary decision outcome

Primary decision outcomein total sample: positive:19; negative: 46

Review decision outcome in totalsample: positive: 10; negative: 34

Total number of primary decisionsobtained for current analysis

Total number of review decisionsobtained for current analysis

45 Decisions (68%) (52 personsincluding 7 dependents)

26 Decisions (59%) (28 personsincluding 2 dependents)

Primary decision outcome inobtained sample: positive:8; negative: 37

Decision outcome in obtainedsample: positive 3; negative 23

21 Decisions (32%) not included 18 Decisions (41%) not included

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115108

A key question that arises in the refugee mental health forensicfield is whether the provision of psychological evidence is aneffective strategy in assisting refugee decision makers to accountfor mental health presentations. Despite the increasing recognitionof the importance of mental health evidence within the refugeedecision-making context, little is known about the use and un-derstanding of such evidence by decision makers. Research fromthe US suggests that the provision of medico-legal and psycho-logical assessments that document injuries and psychiatricmorbidity consistent with torture or other trauma is associatedwith higher asylum acceptance rates compared to thosewho do notreceive such reports (Asgary, Metalios, Smith, & Paccione, 2006;Lustig, Kureshi, Delucchi, Lacopino, & Morse, 2008). The currentstudy is the first to examine the use and understanding of psy-chological evidence by decision makers, whether and how theyincorporate such evidence into their decisions across the primaryand review levels in Australia, drawing on a cohort of asylumseekers assessed prior to lodging their initial applications forrefugee protection.

Methods

Design

We used data from a cohort of asylum seekers making their firstrefugee application in New South Wales, Australia (Silove et al.,2007). All applicants were recruited prior to lodging their initialasylum applications to the Department of Immigration and Citi-zenship (DIAC). Detailed psychological reports were submitted todecision makers at the primary and review decision-making levelson the basis of psychological assessments undertaken with theapplicants. The study was undertaken between 2001 and 2003.

Participants

The sample included a representative cohort of 73 applicantswho filed an asylum application at the primary level at the time ofthe study. The applicants were selected using a cluster-probabilisticsampling method. It was estimated that over 90% of all asylumapplications in New South Wales at the time were lodged inconsultation with a migration representative who assisted theasylum seekers in preparing documents and filing for a protectionvisa. We randomly approached 87 migration agents who had rep-resented asylum seekers during the previous 12-month periodfrom a comprehensive list of 1001 registered migration agents in

New South Wales. Across the recruitment period (2001e2002), 13of these agents assisted a total of 73 applicants (with the others notrepresenting new applicants during the recruitment period), whoall agreed to participate in the study. Of the 73 applicants, 68 wereinformed of the outcomes of their applications after a 3-yearperiod.

During the course of the study, there were a total of 110 refugeeprotection decisions made for the asylum cohort (primary¼ 66;review¼ 44 decisions). Of these, 45 (68%) decision records at theprimary level, and 26 (59%) at the review level were able to beobtained for the current analysis. The 45 decisions records at theprimary level represented 52 persons, 7 of whomwere dependentsof the applicants. At the review level, the 26 (of the initial 45 pri-mary decisions) decisions represented 28 persons, 2 of whomweredependents of the applicants (see Table 1). We refer to cases (de-cisions) (n¼ 45 at the primary level or n¼ 26 at the review level) orapplicants (n¼ 52 at the primary level or n¼ 28 at the review level)respectively as the two units of analysis throughout the article.

Measures

The Structured Clinical Interview for DSM-IV (SCID) (Spitzer,Williams, Gibbon, & First, 1992) is a clinician-administered semi-structured interview to assess current Axis I disorders in accor-dance with DSM-IV. Psychometric studies have yielded soundreliability (>.80), sensitivity (>.75), and specificity (>.80) acrossclinic and community samples. We used the PTSD and majordepression modules.

The Harvard Trauma Questionnaire (HTQ) (Mollica et al., 1992),an internationally recognized measure of PTSD, was used to assessexposure to refugee-related traumas and associated posttraumaticDSM-IV-defined symptomatology. The HTQ demonstrates soundreliability: inter-rater reliability for all events (kappa¼ .93), testeretest reliability (r¼ .89), internal scale consistency (Cronbachalpha¼ .90) as well as sensitivity (78%) and specificity (65%). TheHTQ PTSD scale generates a continuous score based on the totalscore that is the sum of all scores of the items answered divided bythe number of items answered.

The Hopkins Symptoms Checklist (HSCL-25) (Mollica, Wyshak,de Marneffe, Khuon, & Lavelle, 1987) is a 25-item cross-culturallyvalidated measure of depression and anxiety-related symptom-atology, with sound testeretest reliability (r¼ .89 for the entirescale and r¼ .82 for each subscale) and validity (88% sensitivity, 73%specificity). Each item is rated within four categories of responseranging from “not at all” (1) to “extremely” (4). The depressionsubscale was used in the study. The HSCL-25 generates a contin-uous score based on the sum of all scores of the items answereddivided by the number of items answered.

Procedures

The assessments were conducted by psychologists with theassistance of accredited interpreters. Each interview lasted a min-imum of 2 h with some extending across multiple sessions. Base-line interviews were undertaken prior to the primary hearings forall participants following a mean period of 5-monthresidency(M¼ 4.7, SD¼ 16.4) in Australia. Participants were provided witha psychological report following participation in the baselineinterview. The report included findings from diagnostic interviewsand self-report measures and documentation of trauma histories.All reports were prepared by the assessing psychologists under thesupervision of a senior clinical psychologist (ZS, VM). The keyobjective of the report was to document previous traumaticexposure and associated mental health difficulties that wereconsidered to be of relevance to the decision-making context.

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115 109

Collateral information provided by migration representatives wasincorporated into the psychological reports where relevant.

Themeanwaiting time from baseline psychological assessmentsto primary decision was 3.8 months (SD¼ 4.5). For applicants whofiled an appeal for their applications, a second assessment wasundertaken in order to provide an updated psychological reportaccompanying other documents for review submitted to the deci-sion maker. The study was approved by the Human Ethics ResearchCommittee at the University of New South Wales. Applicants in allcases gave consent prior to submitting their reports to the immi-gration officials.

Analysis

We applied a mixed-method approach to analysis. We firstreport descriptive data followed by findings from qualitative andstatistical analysis. In qualitative analysis, we refer to specific ex-amples to illustrate how decision makers made reference to keymental health issues associated with traumatic stress and howthese issues interacted with the psychological reports in theirdecisions.

Qualitative analysisWe applied the constant-comparative method (Strauss &

Corbin, 1998) to analyze qualitative data. The data included: 1)the written statement submitted by the applicant in support oftheir refugee protection claim in which they outline the basis fortheir fear of persecution in their country of origin; 2) the psycho-logical report(s) prepared by the research team; and 3) the primaryand review decision record, a written document prepared by therefugee decision maker providing a detailed outline of the appli-cant’s claim and the basis of the decision to grant or to refuse togrant the applicant a protection visa in Australia. The analysiscommenced using open coding of 10 cases to develop a codingstructure based on key variables ordered conceptually. We focusedon key mental health issues associated with traumatic stressidentified in the psychological reports and how they were under-stood and addressed in the decision records. The first author andothers then applied the final coding framework to the remainingcases using FileMaker Pro 10. Inter-rater reliability of the codingframework was assessed by independent coding of a subset of 10cases. The archival analysis was completed in 2010.

Statistical analysisStatistical analysis was implemented using IBM SPSS 20. t-Test

and chi-square analyses were undertaken to firstly examine dif-ferences in decision outcome of those applicants included (n¼ 52)in the current review and those that could not be included (n¼ 21)due to difficulties in obtaining decision records. Secondly, weexamined differences in diagnoses of PTSD and depression, previ-ous trauma exposure, as well as decision outcome between thosewho were identified in the psychological reports as experiencingthe four mental health issues that might interfere with their abilityto present their case and thosewho did not experience these issues.

Results

Demographic characteristics

The 52 applicants with full decision records included 34 malesand 18 females with a mean age of 39 years (SD¼ 13.5). Applicantsoriginated from 18 countries including Iran (n¼ 16, 31%), Ghana(n¼ 8, 15%), Zimbabwe (n¼ 5, 10%), Afghanistan (n¼ 4, 8%), and thePeople’s Republic of China (n¼ 4, 8%). Themean length of residencyin Australia was 4.7 months (SD¼ 16.4) at baseline assessment.

Prevalence of PTSD and depression

At baseline, 60% (n¼ 31) of applicants met SCID-IV diagnosticcriteria for PTSD and 58% (n¼ 30), criteria for major depressivedisorder (MDD). Previous exposure to potentially traumatic expe-riences according to the HTQ was high with an average endorse-ment of 7 out of 16 trauma categories (M¼ 7.0, SD¼ 3.11) and with56% of participants reporting previous experience of torture.

Asylum decisions

Of the 45 cases reviewed, 8 were successful and granted asylumat the primary level. Of the 37 unsuccessful cases, 35 appealed andrequested a review of their applications. Of these, we were able togain access to 26 decision records from the review tribunal. A re-view of those decisions indicated that 3 cases were granted asylumand 23 were unsuccessful on appeal. The available decisions (pri-mary¼ 45; review¼ 26) included in the analysis represented 52persons at the primary level and 28 persons at the review level (seeTable 1).

The mean acceptance rate for onshore asylum seekers (whoapplied for asylum following their arrival in Australia) at the time ofthe study was 25%, which is broadly representative of the 30%acceptance rate obtained in the current cohort (see Table 1), withthe majority of the positive decisions occurring at the primary level(18%). At the primary level, 6 of the 8 applicants who had theirrefugee claims accepted originated from Iran, with the remainderfrom Romania and Turkey. At the review level, the 3 applicants whowere successful in their refugee claims originated from the China,Zimbabwe, and Aceh.

At the time the study was undertaken, decision makers inAustralia only routinely made available the written decision recordto the applicant if they were not granted a refugee protection visa.In order to access positive decision records it was necessary tomake use of Australia’s Freedom of Information provisions, whichhad to be made individually on behalf of each successful applicantand was not always possible to achieve.

Decision records that were able to be obtained (n¼ 45/66) at theprimary level had more negative (n¼ 37, 82%) than positive out-comes (n¼ 8, 18%) (P¼ .003, Fisher’s exact) compared to those thatwe were unable to obtain (n¼ 21: positive, n¼ 15, 71%; negative,n¼ 6, 29%; P¼ .003, Fisher’s exact). At the review level (n¼ 26/44),the decisions reviewed (n¼ 26) had significantly more negative(n¼ 23, 88%) than positive (n¼ 3, 12%) outcomes (P¼ .001, Fisher’sexact) than those whose decisions wewere unable to obtain (n¼ 9;5 positive and 4 negative decisions) resulting in an overall under-enumeration of positive decision record outcomes at both theprimary and review levels.

There were no significant differences in symptoms of PTSD anddepression at baseline between the applicants included in the re-view (PTSD:M¼ 2.58, SD¼ .67; depression:M¼ 2.45, SD¼ .70) andthose who could not be included due to their decisions being un-available (PTSD: M¼ 2.26, SD¼ .96; t(71)¼ 1.624, P¼ .109;depression: M¼ 2.18, SD¼ .85; t(70)¼ 1.439, P¼ .155).

Psychological evidence and asylum decisions at the primary andreview level

At the primary level, decision makers in 10 out of 45 decisions(22%) referred to the psychological reports in their decisions, withthe remaining 35 making no mention of the psychological evi-dence. Amongst the 10 decisionmakers who referred directly to thepsychological reports, 2 considered the importance of the mentalhealth findings in their decisions to accept the applicants’ asylumclaims, the remaining 8 were critical of the evidence submitted.

Table 2Mental health issues presented in the psychological reports, consideration of these issues by decision makers, and association with overall acceptance rate.

Mental health issuespresented inpsychological reports

Psychiatric diagnoses Prevalence and presentation Reports considered or not bydecision makers at primary andreview level

Interaction between mental healthissues (present vs. absent in thereports) and overall acceptance rate

1. Re-experiencing andavoidance(n¼ 31/52, 60%)

PTSD positive(n¼ 24, 77%)Depression positive(n¼ 21, 68%)

Intrusive thoughts (n¼ 21, 68%);nightmares (n¼ 20, 65%);flashbacks (n¼ 10, 32%);avoidance (n¼ 21, 68%)

Primary:Reports considered (n¼ 9/28, 32%)Reports not considered (n¼ 19/28, 68%)Review:Reports considered (n¼ 11/11)

Present:Asylum granted (n¼ 3/31,10%; P¼ .003)Absent:Asylum granted (n¼ 10/21,48%, P¼ .003)

2. Memory impairment(n¼ 29/52, 63%)

PTSD positive(n¼ 20, 69%)Depression positive(n¼ 20, 69%)

Re-experiencing andavoidance (n¼ 25, 86%);Dissociative symptoms(n¼ 12, 41%)

Primary:Reports considered (n¼ 8/27, 30%)Reports not considered (n¼ 19/27, 70%)Review:Reports considered (n¼ 9/19, 47%)Reports not considered(n¼ 10/19, 53%)

Present:Asylum granted (n¼ 3/29,10%; P¼ .009)Absent:Asylum granted (n¼ 10/23,44%, P¼ .009)

3. Dissociative reactions(n¼ 15/52, 29%)

PTSD positive(n¼ 20, 69%)Depression positive(n¼ 20, 69%)

Emotional numbing anddetachment (n¼ 14, 93%);Depersonalization (n¼ 2, 13%);Dissociative amnesia (n¼ 1, 7%)

Primary:Reports considered (n¼ 4/14, 29%)Reports not considered (n¼ 10/14, 71%)Review:Reports considered (n¼ 8/8)

Present:Asylum granted (n¼ 0/15)Absent:Asylum granted (n¼ 8/37,22%, P¼ .087)

4. Difficulties in disclosingpast experiences(n¼ 10/52, 19%)

PTSD positive(n¼ 20, 69%)Depression positive(n¼ 20, 69%)

Delayed or partialdisclosure (n¼ 10/52, 19%)

Primary:Reports considered (n¼ 6/10, 60%)Reports not considered (n¼ 4/10, 40%)Review:Reports considered (n¼ 2/10, 20%)Reports not considered (n¼ 8/10, 80%)

Present:Asylum granted (n¼ 2/10)Absent:Asylum granted (n¼ 11/42,26%, P> .1)

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115110

At the review level, decision makers in 12 out of 26 decisions(46%) referred to the psychological reports. Of the 12 decisionmakers who referred to the expert reports, 3 in their decisions tooverturn the primary decision findings took a positive view of thepsychological reports, while the remaining 9 who upheld the pri-mary decision findings to refuse the grant of a protection visa madecritical comments on the psychological evidence.

The impact of mental health issues on asylum decision making

From the qualitative analysis we identified four mental healthissues associated with posttraumatic stress as having an impact onasylum decision making: namely, a) re-experiencing and avoidanceof traumatic memories, b) memory impairment, c) dissociativereactions, and d) difficulties in disclosing past experiences. For eachof these issues, we report findings from qualitative and descriptiveanalysis of 71 decisions (primary: 45, review: 26), representing 52asylum applicants (Table 2).

Re-experiencing and avoidanceAccording to the psychological reports, 31 (60%) out of 52 ap-

plicants (representing 28 cases/decisions) exhibited re-experiencing and avoidance symptoms, including re-experiencingof traumatic events in the form of nightmares (n¼ 20, 65%),intrusive thoughts (n¼ 12, 39%), and flashbacks (n¼ 10, 32%), aswell as avoiding thoughts about the traumatic events (n¼ 21, 68%).Of these, 24 (77%) met criteria for PTSD and 21 (68%) for depression.Those who did not report these symptoms had significantly lowerrates of PTSD (n¼ 10/21, 48%; P¼ .039, Fisher’s exact test), but notdepression (n¼ 13/21, 62%; P¼ .162, Fisher’s exact test).

Mr K. The case of Mr K illustrates the impact of re-experiencing andavoidance. The initial report stated:

“[Mr K] explained that he continually tries not to think of histraumatic experiences particularly his witnessing of the helicopterattack on women and children in [country of origin]. However, hereports persistent intrusive unwanted memories and dissociation,

stating that he felt that although his body was in the interviewroom, his soul is still in country of origin with the dead (p. 6)”.

The impact of these symptoms on the capacity of Mr K wasnoted in the psychological report: “[Mr K] appeared to be activelyavoiding the line of questioning. and repeatedly related indirectlyrelated events”. The psychologist advised that the primary decisionmaker consider the impact of the applicant’s prior history and hisposttraumatic symptomatology on his testimony particularly whenassessing his credibility. Mr K’s application was rejected with thedecision maker citing evidence of inaccuracies in his accountsacross repeated interviews. In this case, there was no reference tothe psychological reports that identified similar symptoms andpresented an alternative psychological explanation.

The decision makers in the majority of cases involving re-experiencing and avoidance (n¼ 19, 68%) at the primary level didnot refer to the psychological reports or the applicants’ mentalhealth in reaching their decisions. In 9 (32%) cases, the decisionmakers reacted negatively to the reports, including expressingskepticism about the value of psychological evidence, and rejectingexpert opinion concerning the mental health impact of post-traumatic stress. All applicants with reports documenting re-experiencing and avoidance symptoms were refused asylum atthe primary level.

At the review level, 11 cases of re-experiencing were identifiedin the psychological reports. The decision makers in 2 (11%) casestook note of the value of the psychological evidence in assistingthem to understand the impact of adverse life experiences on themental health of the applicant, with one decision-maker applyingthe expert findings in the final decision. The decision makers in 5(26%) cases were critical of the psychological reports that citedinterference with the applicants’ testimonies as a result of re-experiencing and avoidance. Furthermore, they expressedconcern about the objectivity of the reports as they failed to addresswhat the decision makers regarded to be the central issues in eachcase. In the other 4 (21%) cases, the decisionmakers chose to rely ontheir own observations instead of the expert evidence in assessingthe applicants. The applicants in 3 of the 10 review cases weregranted asylum.

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115 111

Across the primary and review levels, the applicants who re-ported re-experiencing and avoidance were granted asylum at asignificantly lower rate (n¼ 3/31, 10%; P¼ .003, Fisher’s exact test)than those who did not report these symptoms (n¼ 10/21, 48%;P¼ .003, Fisher’s exact test).

Memory impairmentThe psychological reports indicate that 29 (63%) out of 52 ap-

plicants reported ongoing memory impairment. Of these, 20 (69%)met criteria for PTSD and depression at baseline. Those who re-ported memory impairment for the most part also reported re-experiencing and avoidance symptoms (n¼ 25/29, 86%), and to alesser extent dissociative reactions (n¼ 12/29, 41%). The presenceof memory impairment was not statistically associated with PTSD(present¼ 20, 69%; absent¼ 9, 31%; P¼ .571, Fisher’s exact test) ordepression (present¼ 20, 69%; absent¼ 9, 31%; P¼ .571, Fisher’sexact test).

Mr A. The case of Mr A illustrates the impact of memory impair-ment and associated difficulties in recalling and narrating acoherent account of experience. The psychological report statedthat the applicant experienced multiple traumatic events prior toseeking asylum in Australia, including physical assaults, solitaryconfinement, and witnessing rape and other forms of torture. Thereport also documented that the applicant expressed concern abouthis “decaying memories” and his ability to recall details of partic-ular events and people. The psychologist noted that such memoryimpairment presented by the applicant was consistent with thetraumatic event that he reported to have experienced.

In refusing the applicant asylum, the primary decision makerstated that:

“Whilst I can accept claims of depression due to his separation fromhis family I found that the applicant was able to recall numerousdetails in support of his claims at interview and through his writtenstatements. I therefore consider that his mental situation has notprevented him from presenting his claims advantageously for aprotection visa (p. 15)”.

At the primary level, the decision makers in the majority of cases(19/27, 70%) in which memory impairment was identified in thepsychological reports did not refer to the mental health findings inthe reports. Of the 8 decision makers who referred to the reports, 5commented negatively about the psychological evidence, 2 excludedthe expert opinion by relying on their ownobservations tomake thatdecision, and 1 explicitly stated that the report was unhelpful.

“Despite comments .. that [the applicant] was incapable of pre-senting her claims, due to forgetfulness, age, physical and mentalimpairment, the applicant was articulate and lucid in describingher circumstances in [her country of origin]. She appeared to un-derstand the Tribunal’s questions and she was able to provide ameaningful account of her difficulties with the authorities in [hercountry of origin] (p. 12)”.

At the review level, in nearly half of the cases (10/19, 53%) inwhich memory impairment was noted in the psychological reports,the decision makers did not make reference to the findings in thosereports. The decision makers in the other 9 cases in which theapplicants reported memory impairment had a mixed response.Two stated that the reports were useful, 3 made negative com-ments about the reports, 3 substituted their own observations inplace of the psychological evidence, and one stated that the reportwas unhelpful.

Across the primary and review level, the applicants who re-portedmemory disturbanceswere granted asylum at a significantly

lower rate (n¼ 3/29, 10%; P¼ .009, Fisher’s exact test) than thosewho did not report this issue (n¼ 10, 44%; P¼ .009, Fisher’s exacttest).

Dissociative reactionsAccording to the psychological reports, 15 (29%) of 52 applicants

exhibited dissociative phenomena as a result of previous traumaticexposure involving sexual and physical assaults, harassment, andimprisonment. Of these, 14 (93%) reported emotional numbing anddetachment, 2(13%) reported depersonalization, and 1 (7%) disso-ciative amnesia following the traumatic events.

Thirteen (87%) and 12 (80%) out of those reported dissociativereactions met criteria for PTSD and depression at baseline. Thepresence of dissociation was not associated with differences be-tween those diagnosed with PTSD (n¼ 13/15, 87%, P¼ .055, Fisher’sexact test) and depression (n¼ 12/15, 80%, P¼ .208, Fisher’s exacttest) and those who did not (PTSD: n¼ 2/15,13%; depression: n¼ 3/15, 20%).

Mr S. The case of Mr S illustrates the impact of dissociative pre-sentations associated with difficulties in narrating traumatic eventsduring the hearing. Mr S claimed asylum on the basis of his expe-rience of political persecution in his home country. The psycho-logical report submitted to the review decision maker made note ofthe dissociated presentation of Mr S during the interview.Furthermore, the psychologists acknowledged that the symptomsreported by the applicant including affective restrictions, feelings ofshame and humiliation, avoidance were consistent with his allegedexperience of being sexually assaulted en route to Australia. Thepsychologist advised the review decision maker:

“[Mr S’s] [current] psychological state is likely to affect his ability toanswer questions at his interview with the RRT. in particular, hisapparent emotional detachment and passive way of describingpainful aspects of his flight from . and [his] journey to Australiashould not be interpreted as lacking in content or depth, as theblunted affect displayed by [Mr S] is consistent with a pattern ofdiminished responsiveness and withdrawal in persons who havesurvived violent or life-threatening traumas (p. 8).”

In considering this evidence, the decision maker stated that:

“[The] psychologist reported that the applicant was suffering fromposttraumatic stress disorder and depression and that this psy-chological state was likely to affect his ability to answer questionsat an RRT hearing.. [Nevertheless] [Mr S] did not display anydifficulty in understanding or answering questions. . He[appeared] alert, engaged, and is clearly an intelligent man. I do notaccept that he had any difficulty in understanding proceedings oranswering questions. (p. 15)”.

In a negative finding of the credibility of the applicant’s asylumclaim, the decision maker stated that the applicant “set out delib-erately to mislead the Tribunal” and that his claims could not beaccepted as truthful. The decision maker also rejected the expertevidence on the basis that such evidence was based on the appli-cant’s prior histories and his attempts “to mislead the Tribunal”.

The tendency for decision makers to rely on their own obser-vations over the psychological evidence was evident in 3 othercases involving decision makers at the primary and review level inwhich dissociation featured prominently in the mental health ofthe applicants.

At the primary level, the decision makers in 4 out of 14 cases(29%) in which dissociation was cited in the psychological reportswere critical of the evidence in their comments. Of these, 3 regar-ded the reports to be of little value and lacking in objectivity, 1

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115112

chose to substitute their own judgment in assessing the mentalhealth of the applicant. The decision makers in the remaining 10cases made no reference to the psychological evidence.

At the review level, the decision makers in 8 of the appeal casesin which specific features of dissociation were identified gavemixed feedback about the psychological reports. The decisionmakers in 3 cases drew on the mental health findings to supporttheir findings of fact in relation to the applicants. The decisionmakers in 3 cases made skeptical comments about the expert re-ports, stating that they failed to clarify the key issues in question. Inthe other 2 cases, the decision makers relied on their ownreasoning and observations of the applicants in their assessments.

At the primary level, all 15 cases with the presence of disso-ciative symptoms were not granted a protection visa compared to 8out of 37 of non-dissociative cases, a finding that approached sta-tistical significance (P¼ .087). Therewas a higher acceptance rate atthe review level so that overall there were similar acceptance ratesfor applicants with dissociative reactions (n¼ 4/15, 26%; P¼ 1.000,Fisher’s exact test) and those without (n¼ 9/37, 24%; P¼ 1.000,Fisher’s exact test).

Difficulties in disclosing important information about pastexperiences

We identified difficulties amongst the applicants in disclosingpast experiences including delayed disclosure where an applicantfailed to mention details or provide important information relatedto their claims at the primary level, and partial disclosure where anapplicant made incomplete disclosure of such information. Forexample, an applicant may have alluded to being sexually assaultedby having disclosed partially about the assault, but feelings ofshame and humiliation may result in the applicant refraining fromelaboration or even recanting previous statements about theincident.

According to the psychological reports, 10 (19%) out of 52 ap-plicants failed to disclose or made partial disclosure of their pastexperiences at the primary level. Of these, applicants were involvedin partial disclosure of sexual violence, including rape (n¼ 3, 30%),torture during detention (n¼ 4, 40%), and sexual harassment andintimidation (n¼ 1, 10%) by government authorities in their coun-tries of origin. The applicants in the remaining cases failed todisclose completely, with one applicant failing to mention hisoverseas trip to a foreign country prior to claiming asylum, and theother failing to mention a state-sanctioned arrest warrant againsthim in his home country. The reasons provided by those who failedto disclose (n¼ 2, 20%) and who made partial disclosure (n¼ 2,20%) included feelings of shame and humiliation related to thereported experiences. The remaining applicants did not give rea-sons for their delayed or incomplete disclosure in the psychologicalreports.

Eight of the 10 applicants involved in partial or delayed disclo-sure met criteria for PTSD, with 9 meeting criteria for majordepression. The presence of partial or delayed disclosure was notassociated with differences between those who reported PTSD(n¼ 8/10, 80%, P¼ .462, Fisher’s exact test) and depression (n¼ 9/10, 90%, P¼ .136, Fisher’s exact test) and those who did not (PTSD:n¼ 2/10, 20%; depression: n¼ 1/10, 10%). Those involved in partialor delayed disclosure reported a significantly higher trauma count(M¼ 9.40, SD¼ 2.46; t(50)¼�2.91, P¼ .005) than those who didnot (M¼ 6.43, SD¼ 2.99).

The case of Ms B illustrates the challenges to disclosing priorexperiences of persecution in entirety and the impact on the de-cision maker’s evaluation of her credibility.

Ms B. Ms B reported sexual assaults by members of a militia groupin two separation locations (detention and her own home) in her

country of origin. The applicant referred briefly to these incidents,although the sexual nature of the events was not disclosed ordocumented in the personal statement accompanying her primaryapplication. The applicant, after being refused asylum at the pri-mary level, revealed in the follow-up psychological assessment thefull extent of the assaults. Furthermore, she reported feelings ofshame and humiliation as an impediment to disclosing fully detailsof the assaults, and that none of her legal or migration represen-tatives inquired further into these incidents.

The follow-up psychological report stated that:

“The fact that Ms B did not provide details of the alleged sexualassault in her primary application statement or in the supportingdocuments submitted to the Tribunal [at appeal] is entirelyconsistent with individuals experiencing post-traumatic reactions. it is well documented that refugee women who have been rapedor sexually assaulted do not easily disclose their experiencesbecause of the shame and stigma attached to these crimes. (p. 8)”.

The review decision maker recognized the discrepancies in theaccounts of these experiences by the applicants across repeatedinterviews, stating that:

“The Tribunal pointed out to [Ms B] that no mention of these in-cidents [sexual assaults] had been made in the [psychological]report which had been sent to the migration agent. [Ms B]acknowledged that she had not mentioned these incidents to the[psychologist]”..: “I accept that this incident of sexual molestationas described by [the applicant] to the Tribunal (and to the psy-chologists) in great detail would have been distressing andhumiliating for the Applicant. (p. 15)”.

The report also linked the alleged assault to the applicant’s claimof persecution on the basis of political activism, the primary reasonfor her seeking asylum.

“I accept that the Applicant was sexually assaulted by militarypersonnel [in detention] after her return to her home country.However, despite the claims made in the Psychologists Reports [sic]about the reason for the molestation she suffered and the causes ofthe Applicant’s current state of depression and her emotional dis-orders. I do not give much evidentiary weight to these as I find thatthey are merely evidence of what the Applicant told the psychol-ogists was the reasons for what occurred, rather than being evi-dence of the facts themselves (p. 16)”.

The decision makers expressed concerns about the reasonsprovided in the psychological reports for partial or delayeddisclosure at the primary (n¼ 6/10) and review level (n¼ 2/10). Inthe remaining cases, the decision makers questioned the appli-cant’s claim itself without reference to the reasons underlyingpartial or incomplete disclosure. The reasons for refusing asylum inthese cases (10 rejected at the primary level, 8 rejected and 2successful at appeal) varied, relating to testimonial alterations,factual inconsistency, contradictory findings of country informa-tion, and risk assessment. Across the primary and review level, theapplicants who failed to disclose had a comparable outcome (n¼ 2/10, 20%; P¼ 1.000, Fisher’s exact test) to those who did not reportthis issue (n¼ 11/42, 26%, P¼ 1.000, Fisher’s exact test).

Discussion

This study is the first to examine the use and understanding ofpsychological evidence amongst refugee decision makers, drawingon a representative cohort of asylum seekers whowere granted andrefused asylum. Specifically, we examined applicants’ writtenstatements submitted to support their refugee protection claims,

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115 113

psychological reports prepared for the applicants at the primaryand review levels, and the primary and review decision recordsoutlining the basis of the decision to grant or refuse the applicantsrefugee protection in Australia. Overall, half of the decision makersin the study did not refer to the psychological reports in their de-cision records. It is noteworthy, considering this finding, that in-ternational (IARLJ, 2011) and Australian guidelines (MRT/RRT, 2012)have stressed the importance of asylum decision makers consid-ering medical and psychological evidence in assessing refugeestatus, with the failure to refer to such evidence in overseas juris-dictions potentially constituting a reviewable error (Hunter,Pearson, & San Roque, 2013; Hunter, Pearson, San Roque, & Steel,2013; Hunter, Steel, et al., 2013). In the present study, the deci-sion makers at the review level referred to the psychological evi-dence more frequently (46%) than those at the primary level (22%).This finding may reflect a number of factors, including a highernumber of applications at the primary decision-making level,higher levels of training amongst decision makers at the appeallevel, and the opportunity for tribunal members to review potentialerrors in previous decisions. Considering the elevated burden ofmental disorders amongst conflict-affected populations (Steel et al.,2009), and the complexities of asylum determination, there is apressing need for ongoing specialist training for decision makers inrefugee mental health at all levels. Similarly, mental health pro-fessionals providing psychological or psychiatric evidence forimmigration authorities need to be aware of the general principlesof asylum determination, the specific decision-making frameworkin each setting or country, and the key issues in the presentationand formulation of psychological reports. To this end, the researchteam has prepared a set of guidelines to assist refugee decisionmakers, legal representatives, and mental health professionals inthe preparation and use of psychological evidence within therefugee decision-making setting (Hunter, Pearson, & San Roque,2013; Hunter, Pearson, San Roque, & Steel, 2013; Hunter, Steel,et al., 2013). Further research is needed to examine the effective-ness of psycho-legal training for decisionmakers andmental healthprofessionals in the context of asylum decision making.

The present study identified several ways in which decisionmakers commonly dealt with and addressed the psychological re-ports. Those who responded to the psychological evidence favor-ably acknowledged the congruency between the psychologicalreports presented and the applicants’ symptomatology, and theimpact of posttraumatic stress on the applicants’ ability tocompetently narrate their experiences. Conversely, those who tooka negative view of the psychological evidence were critical of thereports, questioned the objectivity and value of such evidence, andchallenged whether it was relevant to the key issues in the asylumapplications. It is noteworthy that a number of decision makerschose to rely on their own observations and reasoning in assessingthe asylum seekers’ mental health as a substitute for the expertevidence submitted to them. The grounds for such substitutionappear to be the decision makers’ own observations without spe-cific inquiry into mental health symptoms within a context wherethere is a lack of specialist training in mental health, a matter ofconsiderable concern. The observation that the majority of decisionmakers at all levels did not make reference to the psychologicalevidence in their decisions is also a finding that raises substantialconcern. Taken together, the findings underscore the importance ofdecision makers taking active steps in considering the use of psy-chological evidence when assessing potential mental health issuespresented by asylum seekers during the asylum assessment pro-cess. In addition, there is accruing evidence that detention policiescontribute to adverse mental health outcomes amongst trauma-tized asylum seekers who have been in detention for prolongedperiods of time (Steel et al., 2011). As a result of these mental health

issues, asylum seekers are likely to face substantial difficulties inpresenting their claims before immigration authorities.

We identified four mental health issues associated with post-traumatic stress as having an impact on asylum decision making:namely, re-experiencing and avoidance of traumatic memories,dissociative reactions, memory impairment, and difficulties indisclosing important information about past experiences. The caseexamples discussed in the article highlight how each of these fac-tors can affect asylum seekers and their ability to effectively presenttheir case before decision makers. For example, re-experiencingand avoidance of traumatic memories can limit disclosure of in-formation; memory impairment can contribute to difficulties inrecalling and recounting past events; and cultural and psycholog-ical factors can lead to delayed or incomplete disclosure. Further-more, in the current sample, the presence in over 60% of theapplicants of re-experiencing and avoidance, together with mem-ory impairment was associated with a lower asylum acceptancerates. It is important to note that in a previous report from thissample we did not identify an association between the level oftrauma exposure, the prevalence of mental disorder and the asylumapplication outcome (Silove et al., 2006). The current finding sug-gest that rather than conferring a general risk, the association be-tween mental disorder and asylum outcome is mediated by thepresence of specific symptoms such as re-experiencing, avoidance,and memory impairment.

It is possible that those with more severe mental health issuesmight have greater difficulties in reporting their traumatic experi-ences or had presented in a way that led to adverse credibility as-sessments by decision makers. From the small number of decisionmakers who referred to these issues in their decisions, it is notpossible to reach any finding on the specific mechanism by whichthese factors impacted on the decision-making process. Forexample, the applicants with severe PTSD symptoms might havebeen more likely to face substantial challenges in accurately recall-ing and narrating their experiences as a result of narrative disorga-nization of the traumatic memory. This could only be furtherexamined via in-depth analysis of transcripts of actual asylum in-terviews which were not available in the current research.

The presence of dissociation and difficulties in disclosure wasnot directly associated with adverse decision outcomes across theasylum determination process. It is noteworthy, however, that allapplicants who reported dissociation or difficulties in disclosurewere refused asylum at the primary decision-making levelcompared to those who did not report these issues. Several of theseapplicants were then successful in their appeals. This finding mayreflect the role of the psychological reports in highlighting andpresenting plausible explanations for omission of information ordiscrepancies in the accounts of the applicants. It is noteworthythat therewas a higher level of engagement at the review level withthe psychological reports for applicants who reported dissociationor difficulties in disclosure compared to the other mental healthissues examined.

There are several limitations in the study that merit discussion.We documented the association between mental health issues inindividual cases and decision makers’ comments about these is-sues, and overall asylum decisions, across both primary and appeallevels. Caution should be applied in drawing causal inferences fromthe data as it cannot be excluded that other factors not assessedaccounted for the association between mental health symptomsand refugee outcome. It is also not possible to attribute any role tothe psychological evidence in influencing the decision-makingprocess without including a detailed narrative analysis of thehearings to identify the nature of asylum seekers’mental states anddecision makers’ responses. One of the key strengths of the currentanalysis over earlier research was the capacity to include both

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115114

positive and negative refugee decision outcomes (Rousseau et al.,2002). It should be noted, however, that we were able to accessonly 40% of the records of positive asylum decisions as compared to80% of negative decisions in the study cohort, and thereforerecognize that there was an under-enumeration of positive de-cisions in our analysis.

Furthermore, although the present cohort was broadly repre-sentative of asylum seekers in Australia at the time of the study, wewere able to recruit only those who received help from registeredmigration representatives in their asylum applications, whichaccounted for 90% of asylum seekers. The finding that decisionmakers tended to be critical of the psychological reports particu-larly in the context of adverse decisions suggests that a morepositive evaluation of the reports might have been observed if morepositive decisions had been able to be included. It is also importantto note that the decision records reviewed as part of this studywerecompleted across the years 2001e2003. Since this time, there hasbeen an increasing awareness amongst decision makers about theemerging body of mental health research relevant to asylumdetermination (Hunter, Pearson, & San Roque, 2013; Hunter,Pearson, San Roque, & Steel, 2013; Hunter, Steel, et al., 2013).Within the Australian context, the Department of Immigration andCitizenship’s procedural advice manual now outlines the potentialimpact of PTSD on the consistency of recollections, and cautionsdecision makers against making an adverse credibility assessmenton the basis of the applicant displaying an inability to give a“precisely accurate or consistent account of some past event” (DIAC,2012), with similar guidelines also developed at the review level(MRT/RRT, 2012). The extent towhich this guidancemay have led toimproved engagement with expert psychological evidence remainsuncertain.

Conclusion

Although there is growing recognition of the significance ofmental health evidence in the asylum determination process, themajority of refugee decision makers in the current study showedlimited reference to and preparedness tomake use of such evidenceacross the primary and review levels. Furthermore, there was evi-dence to indicate that asylum seekers with high levels of symptomsassociated with PTSD, particularly cognitive intrusion-avoidanceand memory disturbance, were more likely to receive negativerefugee decision outcomes. These findings raise concern thatasylum seekers with significant mental health impairment may beconsistently disadvantaged in the refugee decision-making processin Australia and globally. There is a pressing need for both mentalhealth and legal professionals included in asylum evaluations toimprove their understanding and use of mental health evidence viainterdisciplinary communication, training and collaboration. Inconjunctionwith this analysis we launch a set of guidelines to assistrefugee decision makers, mental health professionals and legaladvisers improve the quality and use of psychological evidencewithin the refugee decision-making context (Hunter, Pearson, &San Roque, 2013; Hunter, Pearson, San Roque, & Steel, 2013;Hunter, Steel, et al., 2013).

References

Asgary, R., Metalios, E., Smith, C., & Paccione, G. (2006). Evaluating asylum seekers/torture survivors in urban primary care: a collaborative approach at the BronxHuman Rights Clinic. Health and Human Rights, 9, 164e179.

Bernsten, D., & Rubin, D. (2007). The centrality of event scale: a measure of inte-grating a trauma into one’s identity and its relation to post-traumatic stressdisorder symptoms. Behavior Research Therapy, 44, 219e231.

Bogner, D., Brewin, C., & Herlihy, J. (2010). Refugees’ experiences of home officeinterviews: a qualitative study on the disclosure of sensitive personal infor-mation. Journal of Ethnic and Migration Studies, 36, 519e535.

Bogner, D., Herlihy, J., & Brewin, C. (2007). Impact of sexual violence on disclosureduring Home Office interview. British Journal of Psychiatry, 191, 75e81.

Brewin, C. (2007). Autobiographical memory for trauma: update on four contro-versies. Memory, 15, 227e248.

Brewin, C. (2011). The nature and significance of memory disturbances in post-traumatic stress disorder. Annual Review of Clinical Psychology, 7, 203e227.

Cohen, J. (2001a). Errors of recall and credibility: can omissions and discrepancies insuccessive statements reasonably be said to undermine credibility of testi-mony? Medico-Legal Journal, 69, 25e34.

Cohen, J. (2001b). Questions of credibility: omissions, discrepancies, and errors ofrecall in the testimony of asylum seekers. International Journal of Refugee Law,13, 293e309.

Dauvergne, C., & Millbank, J. (2003). Burdened by proof: how the Australian refugeereview has failed lesbian and gay asylum seekers. Federal Law Review, 31, 1e44.

DIAC. (2012). Guidelines for assessing credibility. Published by Australia: Departmentof Immigration and Citizenship (DIAC). Retrieved from http://www.immi.gov.au/media/publications.

Ehlers, A., Mayou, R., & Bryant, R. (1998). Psychological predictors of chronicposttraumatic stress disorder after motor vehicle accidents. Journal of AbnormalPsychology, 107, 508e519.

Eytan, A., Laurencon, M., Durieux-Paillard, S., & Ortiz, N. (2008). Consistency of auto-biographical memories in asylum seekers. American Journal of Psychiatry, 165, 6.

Foa, B., Molnar, C., & Cashman, L. (1995). Change in rape narratives during exposuretherapy for posttraumatic stress disorder. Journal of Traumatic Stress, 8, 675e690.

Granhag, P., Strowall, L., & Hartwig, M. (2005). Eyewitness testimony: tracing the be-liefs of Swedish legal professionals. Behavioral Science and the Law, 23, 709e727.

Hamlin, R. (2012). International law and administrative insulation. A comparison ofrefugee status determination regimes in the US, Canada, and Australia. Law andSocial Inquiry, 5, 45e46.

Herlihy, J., Gleeson, K., & Turner, S. (2010). What assumptions about humanbehavior underlie asylum judgments? International Journal of Refugee Law, 22,351e366.

Herlihy, J., Scragg, P., & Turner, S. (2002). Discrepancies in autobiographical mem-ories: implications for the assessment of asylum seekers. Repeated interviewsstudy. British Medical Journal, 19, 324e327.

Herlihy, J., & Turner, S. (2006). Should discrepant accounts given by asylum seekersbe taken as proof of deceit? Torture, 16, 81e92.

Herlihy, J., & Turner, S. (2009). The psychology of seeking protection. InternationalJournal of Refugee Law, 21, 171e192.

Hunter, J., Pearson, L., & San Roque, M. (2013). Mental health expertise in refugeestatus decision-making: judging or caring? The Federal Law Review. underreview.

Hunter, J., Pearson, L., San Roque, M., & Steel, Z. (2013). Asylum adjudication, mentalhealth and credibility evaluation. The Federal Law Review. under review.

Hunter, J., Steel, Z., Pearson, L., San Roque, M., Silove, D., Frommer, N., et al. (2013).Managing and understanding psychological issues among refugee applicants: Re-sources manual and guidelines for best practice. Sydney: Faculty of Law andPsychiatry Research and Teaching Unit, University of New South Wales.Retrieved from http://www.law.unsw.edu.au/sites/law.unsw.edu.au/files/images/unsw_refugee_study_resources_manual_v2_-_final_with_cover_21032013.pdf.

IARLJ. (2011). International Association of Refugee Law Judges’ guidelines on thejudicial approach to expert medical evidence. Published by International Asso-ciation of Refugee Law Judges. Retrieved from http://www.iarlj.org/general/images/stories/working_parties/guidelines/Final_guidelines_March_2011.pdf.

Jelinek, L., Randjbar, S., Seifert, D., Kellner, M., & Moritz, S. (2009). The organizationof autobiographical and non-autobiographical memory in posttraumatic stressdisorder (PTSD). Journal of Abnormal Psychology, 118, 288e298.

Joseph, S., Dalgleish, T., Thrasher, S., Yule, W., Wiliams, R., & Hodgkinson, P. (1996).Chronic emotional processing in survivors of the Herald of Free Enterprisedisaster: the relationship of intrusion and avoidance at 3 years to distress at 5years. Behaviour Research and Therapy, 34, 357e360.

Kagan, M. (2003). Is truth in the eye of the beholder? Objective credibility assess-ment in refugee status determination. Georgetown Immigration Law Journal, 17,367e415.

King, D., King, L., Erickson, D., Huang, M., Sharkansky, E., & Wolfe, J. (2000).Posttraumatic stress disorder and retrospectively reported stressor expo-sure: a longitudinal prediction model. Journal of Abnormal Psychology, 109,624e633.

Kneebone, S. (1998). The refugee review tribunal and the assessment of credibility:an inquisitorial role? Australian Journal of Administrative Law, 5, 78e97.

Kneebone, S. (2003). Believable tales: credibility and proving a “well-founded fearof persecution”. In Migration, mental health, and human rights conference.

Lustig, S., Kureshi, S., Delucchi, K., Lacopino, V., & Morse, S. (2008). Asylum grantrates following medical evaluations of maltreatment among political asylumapplicants in the United States. Journal of Immigration and Minority, 10, 7e15.

Macklin. (2006). The truth about credibility. Conference paper. Paper presented at theInternational Association of Refugee Judges.

Masinda, M. (2004). Quality of memory: impact on refugee hearing decisions.Traumatology, 10, 131e139.

McNally, R. (2003). Remembering trauma. Cambridge, MA: Harvard University Press.

K. Tay et al. / Social Science & Medicine 98 (2013) 106e115 115

McNally, R., Lasko, N., Macklin, M., & Pitman, R. (1995). Autobiographical memorydisturbance in combat-related posttraumatic stress disorder. Behavior ResearchTherapy, 33, 619e630.

Meffert, S., Musalo, K., McNiel, D., & Binder, R. (2001). The role of mental healthprofessionals in political asylum processing. Journal of American Academy ofPsychiatry and Law, 38, 479e489.

Millbank, J. (2009). The ring of truth: a case study of credibility: assessment inparticular social group refugee determinations. International Journal of RefugeeLaw, 21, 1e33.

Mollica, R., Caspi-Yavin, Y., Bollini, O., Truong, T., Tor, S., & Lavelle, J. (1992). TheHarvard trauma questionnaire: validating a cross-cultural instrument formeasuring torture, trauma, and posttraumatic stress disorder in Indochineserefugees. Journal of Nervous Mental Disease, 180, 111e116.

Mollica, R., Wyshak, G., de Marneffe, D., Khuon, F., & Lavelle, J. (1987). Indochineseversions of the Hopkins symptom checklistd25: a screening instrument for thepsychiatric care of refugees. American Journal of Psychiatry, 144, 497e500.

Moradi, A., Herlihy, J., Yasseri, G., Shahraray, M., Turner, S., & Dalgleish. (2008).Specificity of episodic and semantic aspects of autobiographical memory inrelation to symptoms of posttraumatic stress disorder (PTSD). Acta Psychologica,127, 645e653.

MRT/RRT. (2012). Guidance on the assessment of credibility. Published by MigrationReview Tribunal/Refugee Review Tribunal. Retrieved from http://www.mrt-rrt.gov.au/CMSPages/GetFile.aspx?guid¼df229896-7a1e-49c9-aa88-f42feeed0cde.

Porter, S., & Birt, A. (2001). Is traumatic memory special? A comparison of traumaticmemory characteristics with memory for other emotional life experiences.Applied Cognitive Psychology, 15, 101e117.

Prabhu, M., & Baranoski, M. (2012). Forensic mental health professionals in theimmigration process. Psychiatry Clinics of North America, 35, 929e946.

Robinson, W. C. (1998). Terms of refuge: The Indochinese Exodus and the internationalresponse. London: Zed Books.

Rousseau, C., Crepeau, F., Foxen, P., & Houle, F. (2002). The complexity of deter-mining refugeehood: a multidisciplinary analysis of the decision-making pro-cess of the Canadian immigration and refugee board. Journal of Refugee Studies,15, 43e70.

Silove, D., Steel, Z., Susljik, I., Frommer, N., Loneragan, C., Brooks, R., et al. (2006).Torture, mental health status and the outcomes of refugee applications among

recently arrived asylum seekers in Australia. International Journal of Migration.Health and Social Care, 2(1), 4e14.

Silove, D., Steel, Z., Susljik, I., Frommer, N., Loneragan, C., Chey, T., et al. (2007). Theimpact of the refugee decision on the trajectory of PTSD, anxiety, and depres-sive symptoms among asylum seekers: a longitudinal study. American Journal ofDisaster Medicine, 2(6), 321e329.

Southwick, S., Morgan, A., Nicolaou, A., & Charney, D. (1997). Consistency ofmemory for combat-related traumatic events in veterans of Operations DesertStorm. American Journal of Psychiatry, 154, 173e177.

Spitzer, R., Williams, J., Gibbon, B., & First, M. (1992). The structured clinical inter-view for DSM-III-R (SCID): history, rationale, and description. Archives of GeneralPsychiatry, 49, 624e629.

Steel, Z., Chey, T., Silove, D., Marnane, C., Bryant, R., & van Ommeren, M.(2009). Association of torture and other potentially traumatic eventswith mental health outcomes among populations exposed to mass con-flict and displacement. Journal of the American Medical Association,302(5), 537e549.

Steel, Z., Frommer, N., & Silove, D. (2004). Part 1dthe mental health impacts ofmigration: the law and its effects failing to understand: refugee determinationand the traumatized applicant. International Journal of Law and Psychiatry, 27(6),511e528.

Steel, Z., Momartin, S., Silove, D., Coello, M., Aroche, J., & Tay, K. (2011). Two-yearpsychosocial and mental health outcomes for refugees subjected to restric-tive or supportive immigration policies. Social Science & Medicine, 72(2),1149e1156.

Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and pro-cedures for developing grounded theory (2nd ed.). California: Sage.

Turner, S., & Herlihy, J. (2009). Working with refugees and asylum seekers. Psy-chiatry, 8, 322e324.

UNHCR. (2012a). Beyond proof: Credibility assessment in EU asylum systems. Geneva.Retrieved from: http://www.unhcr.org/51a8a08a9.html.

UNHCR. (2012b). DisplacementdThe new 21st century challenge. UNCHR globaltrends 2012. Geneva.

Warring, J., Payne, J., Schacter, D., & Kensinger, E. (2010). Impact of individual dif-ferences upon emotion-induced memory trade-offs. Cognition and Emotion, 24,150e167.