Barriers to Accessing and Negotiating Mental Health Services in Asylum Seeking and Refugee Populations: The Application of the Candidacy Framework

Increasing numbers of people are leaving their homelands because of human rights violations, persecution and conflict. By June 2018, there were an estimated 68.5 million forcibly displaced individuals worldwide of whom 3.1 million were classified as asylum seekers, and 25.4 million as refugees [1]. The arrival of such high numbers of asylum seekers and refugees (AS&R) places substantial pressures on host countries and their services, including mental health (MH) care systems [2, 3].

AS&R can be subject to pronounced stressors and adverse conditions pre-migration, during migration and/or post-migration (i.e. [4‐6]). In light of potential exposure to these stressors, it is perhaps unsurprising that AS&R show higher rates of post-traumatic stress disorder (PTSD) than host populations [5, 7, 8]. Indeed, PTSD rates have been noted to be almost 10 times more frequent in AS&Rs than in age-matched host populations [9]. The literature shows there is high variability in the studied prevalence rates of mental disorders in AS&R populations, compared to host populations. Bogic et al. [7] also found that refugee samples are likely to have high prevalence rates of depression, which often exceed those reported by samples in host nations. However, a review conducted by Priebe et al. [5] concluded that the rates of mood, psychotic, and substance-use disorders found in AS&Rs groups are within the range of the rates present in host groups. Although the evidence base for prevalence rates varies, the literature shows that the exposure to adverse events can have a negative impact on the MH of AS&R. Given the high absolute numbers of AS&Rs moving across borders, this can constitute a significant challenge to healthcare systems in receiving countries.

Research suggests that despite this increased vulnerability, there is often an underrepresentation of AS&Rs in the health care (HC) services [10, 11]. Identified challenges to accessing services include social, linguistic, economic, clinical severity, and cultural differences in symptom presentation, as well as systemic discrimination [10, 12, 13]. There is also evidence that legal entitlement; formal access to care regulations and the migration process inhibit access in various high-income countries [3]. All individuals have a fundamental legal right to health and to access HC, which is represented both in international and European instruments, such as the European Charter of Fundamental Rights [14]. However, depending on migration status, migrants may have limited entitlements to HC due to national laws and policies [15]. For example, the structure of health systems, which is determined by national policies, can determine the availability of services, the need for HC insurance and the extent of HC coverage, amongst others, which can all impact on the ability to access HC in subgroups of migrants [16].

The candidacy framework (CF) was initially developed as a counter to existing ideas of ‘access’ that draw on data about service utilisation (e.g. number of consultations), but which often fail to capture the complex processes involved in navigating care and fails to account for those who do not seek or are refused services. Dixon-Woods and colleagues summarize candidacy as the ways in which eligibility for medical help and intervention is negotiated between individuals and HC services [17]. Candidacy can be understood as a dynamic and contingent process which is constantly defined and redefined through interactions between the individual and professionals. Therefore, people’s previous interactions and experiences with HC services and professionals can also shape an individual’s candidacy [17]. As such, an individual’s identification of their ‘candidacy’ for accessing and negotiating HC services can be culturally, structurally and professionally constructed [18]. This framework provides a means to explore these negotiations and how they can act as barriers to care [17]. The CF proposes seven overlapping stages two of which address immediate access (stage 1 and stage 2) and five which address negotiation (see Table 1).

The CF has thus far mainly been applied in populations whose entitlement to care is relatively stable and comprehensive. Mackenzie et al. have called for the exploration of candidacy in contexts where vulnerable individuals may be subject to compromised services (i.e. lack of citizenship or stigma) [19]. To date, only one study has specifically applied the CF to understanding the help-seeking trajectory of asylum seekers [20]. The study found that asylum seekers’ precarious migratory status constrained their candidacy for obtaining HC. Barriers included having misinformation about HC coverage, tiresome administrative procedures specific to asylum seekers, and long waiting times. The findings showed that migratory status and feelings of marginalization and insecurity that come from their migrant status, appeared to amplify the effects of the barriers to care and even minor difficulties to access could have dramatic effects on future help-seeking behavior [20].

The current review uses the CF to synthesize qualitative research findings investigating barriers to accessing and negotiating MH services for AS&Rs in high-income countries (HIC). The structure and delivery of HC services (including MH services) in HIC are comparatively well resourced and formalized. As such, the exclusive focus on including studies undertaken in HIC in the current review allowed for a fuller examination of barriers and facilitators relating to accessing and negotiating services than including studies conducted in low- and middle-income countries, where services may be non-existent, would have permitted. With HIC-based HC services and providers seeing increasing numbers of AS&R groups [21, 22], there is also need for a more detailed understanding of the barriers to accessing specialist services in HICs. This review is the first to focus specifically on barriers to MH services for AS&R populations by using a CF. The findings of this review can be used to inform the design and delivery of forms of MH support for this underserved population in HIC of resettlement.

The PsychINFO, Medline, Web of Science, SocINDEX and Embase databases were searched up to December 2018. Each search contained three segments; (1) asylum seekers, refugees and displaced persons, (2) MH services and MH problems and (3) candidacy, see “Appendix A” for an example of the full search strategy for the PsycINFO database. The search strategy used was adjusted to each database using the Kings College London library guide [23]. Additionally, reference chaining was completed—a process by which academic papers that have cited an included study are electronically identified and screened for potential inclusion and the reference list of each included study are also searched for studies that could meet eligibility criteria for inclusion.

Two researchers (CB and FR) independently screened the titles and abstracts of all the articles, and the full texts of potentially relevant papers. This gave a moderate inter-rater reliability (κ = 0.42) [24]. Discrepancies were discussed with CM and RW. All qualitative peer-reviewed publications in English exploring barriers faced by adult AS&Rs, or displaced persons to accessing MH services, mental HC delivery, or help-seeking behaviors in HIC were included. Displaced persons were included to ensure all forms of forced displacement were taken into account, including irregular migrants, provided the displacement took place in or to HIC. Books, chapters, dissertations, literature reviews, and theoretical texts were excluded. Articles focusing on individuals under the age of 18 years were also excluded. Studies were also excluded if they did not elicit primary data from participants.

Each paper was individually assessed for quality by author CB using the Critical Appraisal Skills Program (CASP) tool for qualitative studies [25]. A data extraction form was used to summarize bibliographic information, study design, key findings, and limitations. The seven stages of the CF were included in the data extraction process to highlight which study addressed which stage. Author CB read each paper and conducted the data extraction, which was monitored by CM.

A two-stage critical interpretive synthesis (CIS) [17, 26] approach was used. In stage one ‘First order constructs (i.e. direct quotes used in the papers) and second order constructs (i.e. researchers’ interpretations based on existing theories) were identified and merged across studies. This was done by initially extracting all the direct quotes which addressed the themes of accessing and/or negotiating HC from each paper. For example, a first order construct found in a paper published by Teunissen et al. [27] was the quote: ‘Yeah but we didn’t knew that you can go to a GP with depression’ (p. 8). The quotes were put in a table together with the second order constructs provided by the original authors of the study. In this case, Teunissen et al. [27] interpreted the quote as demonstrating a lack of recognition and trust of the GP being a doctor who could treat mental illness. The first and second order constructs were compared and contrasted across the different studies through which third order constructs emerged. In this example, the third order construct was ‘understanding a new system’. This process was undertaken by author CB and peer-reviewed by a second researcher (CM).

In stage 2, evidence from across the studies including first, second and third order constructs were integrated into the synthesizing argument, namely the seven stages of the CF. In this example, the first, second and third order constructs mapped onto stage 2 (Navigation). This was peer reviewed by researcher CM and author RW. Overall, a deductive qualitative approach was used.

The identification of candidacy was dependent on two different third order constructs.

Using services was dependent on three different third order constructs.

Often influenced by the previously mentioned barriers, some individuals delayed seeking medical help until reaching a crisis point and thereby accessing urgent care. This was particularly the case for undocumented migrants. This delay was at times also influenced by previous experiences, where individuals described a substantial gap between initial refusals and further help-seeking behavior. Initial refusals influenced the permeability of services in that participants felt they were not taken seriously when reaching out for help. If past experiences were positive, individuals were more willing to establish their candidacy again.

‘If I get sick I pop pills and wait. And when I say pills I mean over the counter shit, not prescriptions. If it gets really bad then I have to decide if I think I will die. If I think I will, I go to Emergency. If I don’t then I wait in pain. Why do you ask me about family doctors? Walk-in-clinics? Are you kidding? I have no papers.’ [34, p. 171]

Economic worries also restricted permeability, and many individuals described learning about their entitlements on a trial and error basis: “… I do not know [if it will be covered] until I try, when I go” [20, p. 54]. The willingness to do so was described as dependent on whether it was themselves or their children who were in need of care, with there being less hesitation when it concerned a child.

Continuity of care, sub-specialties, and preventive care were largely unknown or unavailable to users. Importantly, they often depended on a range of support services such as non-governmental organizations to access care and be referred to specialist care. Additionally, there also appeared to be cultural norms which influenced whether services were seen as permeable. Evidence showed issues with religion, language, and expectations on what would happen if they did, for example many participants expressed that they believed that going to the doctor is what makes you sick therefore staying away keeps you healthy. These barriers highlight the lack of knowledge and therefore the lack of permeability of specialist services for AS&Rs.

Barriers when appearing at services were linguistic; attitudes and perceived discrimination; and cultural competency.

Findings showed that once AS&Rs had asserted their candidacy by presenting to health services, the professional judgements made regarding their candidacy strongly influenced subsequent access to services. AS&Rs highlighted a lack of resources and inconsistencies between providers. Providers were often perceived as overwhelmed with few options for referring clients who required continuing MH care. Programs designed for AS&Rs often lacked funding or were oversaturated with long waiting times. This resulted in concerns about timely access to a specialist’s opinion. It was however generally recognized that severe conditions were referred more quickly. Respondents also mentioned complex referral processes and eligibility criteria for accessing MH services leading to negative experiences and absence of clear guidance as to how to navigate the system.

“doctors at [A] they suggested [B]. I contacted [B] and then they couldn’t do help much. Then I was transferred to [C] and from there to [D] so it’s like a little tour.” [37, p. 29]

Service users also described experiences of being turned away from the service if they were not assessed to be ill enough when they reached out for help. The lack of referrals led to feelings of rejection by the system and AS&R experienced that they often had to rely on other people to negotiate contact with the services on their behalf in order to be taken seriously.

Cultural concerns with regards to adjudication also emerged. For example, concerns regarding whether HC providers in the country of resettlement were familiar with the common diseases in their countries of origin. Another example that arose was the feeling amongst participants that an illness should be treated in the early stages rather than waiting to see the symptoms develop ‘they let your illness become very old here’ [38, p. 522].This was further underscored by the idea that health services may lack staff that is knowledgeable and sensitive to the particular needs of AS&R populations.

This stage was dominated by concerns over an excessive focus on medication. Experiences of emphasizing watchful waiting approaches and simple self-medication, was perceived by some as revealing a lack of interest in them and their circumstances. Simultaneously, when prescribed medication, there was a lack of understanding what it was for and worries around consequences such as addiction, side effects, or medication leading to worsening of the problem emerged across studies. Mostly, these findings suggested a lack of communication between the provider and patient regarding the purpose of the medication.

‘Sometimes I see these pills… I mean I don’t think that these pills are good. They make me numb. Sometimes I decide to give up. I decide to skip taking them to see what would happen.’ [36, p. 263]

A few individuals in the studies described their statements and behaviors as being exaggerated to fit within a particular illness framework, making it more difficult for them to come to terms with their diagnosis and accept treatment offers. The lack of acknowledgement of the individuals’ perspective towards treatment and their past experiences led to patients feeling detached from their treatment course. In some cases, this resulted in individuals reducing or discontinuing their medication without the involvement of the health professional. This was at times also influenced by the environment including family and clergy, who even if the medications were prescribed sometimes discouraged the individual from taking them. Often this was prioritized over treatment, and the disparity between lay beliefs and Western understanding of MH created a clash of understanding between the service user and the system.

‘If a doctor says you need medication, and the pastor says no. You won’t take it’ [30, p. 212]

Issues around the economic burden of medication also emerged as a reason for resistance and discontinued treatment. Out-of-pocket payment proved to be challenging especially for those who do not have the finances to cover food and housing expenses.

The relationship with the individual provider was highlighted as being essential to the continued use of services. Specifically, trust, rapport and respect in the relationship were key, and suggested that satisfaction with operating conditions and production of candidacy were very person dependent and could take years before they found a provider with whom they developed a deeper connection.

‘Sometimes they were asking very like personal questions that I didn’t like…The journey that we had, like how many days were you in the boat, and I never want to think about it… [later] People are different, like we have saying; ‘‘jungle has dry and wet—some trees are alive, some trees are dead, and they are different.’’ And people are the same; some people like to talk about their selves, their families, and some people want to keep a secret.‘[37, p. 32]

There were also instances in which individuals had not been able to build this relationship which often led to discontinuing the care. The main reason for this was providers not meeting expectations or not being adequately responsive to needs. A need for awareness of the individuals’ cultural background, previous experiences, and understanding how the individual made sense of their illness was very salient. Findings showed the need for practitioners to avoid assumptions and learn from the patient as an individual in order to accommodate nuances in ethnic and religious identities. For example, one female refugee who had previously been incarcerated in Iraq highlighted her distress upon being in a closed ward showing the need for providers to understand the individual’s past: ‘And the door was shut. It was a closed ward. It comes to my mind again, how we were in prison in my country. The door was shut. It was very difficult.’ [39, p. 9749].

Lastly, studies reported accounts of professionals being out of reach due to a lack of time giving rise to feelings of neglect, unworthiness, and frustration in patients. Furthermore, the flexibility of services to respond to individual needs was highly valued. This included adapting the frequency of consultations to personal preferences or maintaining contact with the provider whilst waiting to access specialized treatment.

“he was not helpful or he was just not getting us… I felt he was being disrespectful… we were new to the country and… we had to travel by train then take a tram and sometimes we might be a bit late, but he wasn’t understanding one bit.” [37, p. 35]

The identification of candidacy (stage 1) was dominated by issues relating to the interpretations by AS&Rs of symptoms and social barriers. The data showed that AS&R recognize their symptoms as requiring help, however they often turn to informal services. Previous evidence has suggested that traditional explanatory models of health held by ethnic minority groups can impact on their help seeking behavior from Western HC services. This may be attributable to different explanatory models regarding MH, specifically different holistic beliefs about causality that do not correlate with the western medical model [40]. Certain beliefs about causality can lead directly to shame and stigma, such as MH problems as punishment for wrong-doings [41], as God’s will, and as black-magic, jinn or possession by spirits [42, 43]. On the other hand, in a qualitative study conducted on a Thai Muslim community, family and key stakeholder participants rejected the idea that schizophrenia had stigma since the illness was Allah’s will [44]. Consequently, individuals may choose to access more traditional and faith-based healing practices [45], as was found in the current review. Access to services may not be sufficient, it must be accompanied by efforts to increase MH literacy for communities and training for traditional and faith-based healers to improve referral pathways to formal services and decrease stigma.

The concerns over structural barriers (i.e. fear of financial contribution) and unknown consequences (i.e. legal repercussions) to accessing services found in stage 2 (navigation) suggest that there is a clear need to provide more knowledge on available services and entitlements to care in this population across Western countries. The unknown consequences of accessing services combined with an inherent lack of trust in public organizations and/or fear of being reported to authorities can make it particularly challenging for individuals to trust HC systems especially during the asylum process [46]. Their migratory status has the potential to perpetuate social dependence and economic marginalization [47] and therefore hindering their assertion to candidacy and accessing health services.

The negotiation stages highlighted the dynamic nature of the system and more specifically the constant negotiation between service users and HC providers. Overall, service-level responsiveness was inadequate with waiting lists, eligibility criteria, and continuity of care being described as common and distressing. The findings suggest that power distributions were asymmetrical at times between HC providers and AS&R including the enforcement of dominant values onto services users and perceived discrimination.

Theorists commenting on the difference between illness and diseases have emphasized ‘illness’ as the individual’s lived experience of symptoms and disability; and ‘disease’ as the HC provider’s representation of the disorder after having reworked the person’s account into a medical framework [48, 49]. Understanding how individuals create meaning in their illness can largely influence care and increase diagnostic validity [50, 51]. This calls for the need for culturally competent care, which exists when providers are knowledgeable of the potential and actual factors that can influence their interaction with service users and have training to address the cultural divide [52]. However, dominance of the bio-medical model may fail to adequately acknowledge the social and cultural basis of MH. Providers can be influenced by stereotypes and potentially homogenize this population into a single pathologized identity, or lack training to identify symptoms unique to other cultures [53]. Therefore, providers must constantly reflect on their own values, attitudes and behaviors that could be influencing the relationship and can both directly and indirectly create barriers to care [54‐56].

Language was flagged as a major barrier throughout the current review. In terms of access, individuals were scared providers would not speak their language or understand their symptoms. In terms of negotiating the system, language existed as a barrier throughout the stages. The lack of competent interpretation was said to complicate the encounter and translational services were often not available for comparatively rarely spoken languages and dialects. When this occurred, providers often used family members as translators, which highlight suboptimal standards as this has implications for potential bias in the interpretation, and reduced willingness on the AS&Rs’ behalf to open up. MH providers themselves have also reported similar issues including lack of access to or poor-quality interpretation services in research [57, 58]. This has been found to impact empathetic responses, decrease rapport, service user satisfaction and has shown to increase medical error in previous research [59, 60]. The sensitive nature of AS&Rs’ experiences demands highly competent interpretation services therefore there is a need to train clinicians systematically in the efficient use of interpreters, cultural brokers and cultural formulations as has been highlighted previously [61, 62]. Additionally, interpreters may require additional training to work with AS&R and clinicians in what may be challenging consultations. Piacentini et al. [63] have previously highlighted the need for more training measures that move beyond diversity and/or race awareness, and which use a more holistic approach to understanding how different social identities and multi-dimensional markers of difference come to be produced and reproduced in interpreter-mediated healthcare encounters with migrant populations. They argue that these social identities and markers of difference include language, culture, ethnicity, age, gender, and also immigration status. Therefore, interpreters need to be aware how these variables intersect specifically with language.

The use of the CF as synthesizing argument for CIS has proven to be a useful way to conceptualize barriers and underlying constructs that influence access and negotiation. Using a systematic review to bring this knowledge together has allowed us to cast the net wide and integrate findings from different global settings into new evidence-based knowledge.

Moving forward, a holistic approach incorporating input from a range of stakeholders is needed to address the barriers found in this review, including the work of academics; policy makers and HC providers who all need to acknowledge the impact of country of origin, language, culture and status on MH service provision. Most importantly, the idea that ‘one size does not fit all’ should be at the forefront. Once service users have accessed mainstream health services, simple referral processes and provision of adequate information can facilitate treatment, for example through websites [46, 62]. Furthermore, sensitivity trainings, hiring professionals who share the persons’ ethnicity or language, and improvement of interpretation services are needed. Additionally, interpreters may require additional training to work with AS&R and clinicians in what may be challenging consultations. Piacentini et al. [63] have previously highlighted the need for more training measures that move beyond diversity and/or race awareness, and which use a more holistic approach to understanding how different social identities and multi-dimensional markers of difference come to be produced and reproduced in interpreter-mediated healthcare encounters with migrant populations. They argue that these social identities and markers of difference include language, culture, ethnicity, age, gender, and also immigration status. Therefore, interpreters need to be aware how these variables intersect specifically with language. Most importantly, AS&Rs need to be engaged as stakeholders and stand at the center of finding solutions to achieving accessible and negotiable services.

There is also a need for qualitative research into displaced populations’ barriers to HC in low and middle-income countries. This review focused on high-income countries but can be seen as examples of the types of issues that local MH services should be exploring with their own AS&R communities. The CF has thus far only been used in high-income settings, therefore future research should investigate the suitability of using the CF in low and middle-income settings where more macro level barriers to care may exist. Lastly, this review only considered barriers to access and negotiation rather than including facilitators. There is a need for future reviews to address facilitators that can increase contact with services.