VU Research Portal

Understanding and Treating Suicidal Ideation among Turkish Migrants Eylem, O.

2020

document version Publisher's PDF, also known as Version of record

Link to publication in VU Research Portal

citation for published version (APA) Eylem, O. (2020). Understanding and Treating Suicidal Ideation among Turkish Migrants: Challenges and Innovations.

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal ?

Take down policy If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

E-mail address: [email protected]

Download date: 07. Oct. 2021 Özlem Eylem

UNDERSTANDING AND TREATING SUICIDAL IDEATION AMONG TURKISH MIGRANTS Challenges and Innovations

Understanding and treating sUicidal

ideation among tUrkish migrants

Challenges and Innovations

Özlem Eylem Colofon

Understanding and treating sUicidal ideation among tUrkish migrants - Challenges and Innovations by Özlem Eylem ISBN/EAN: 978-94-028-2118-5

Copyright © 2020 Özlem Eylem All rights reserved. No part of this thesis may be reproduced, stored or transmitted in any way or by any means without the prior permission of the author, or when applicable, of the publishers of the scientific papers.

Cover design by Larisa Wiegant Layout and design by Birgit Vredenburg, persoonlijkproefschrift.nl Printed by Ipskamp Printing | proefschriften.net promotoren: prof.dr. A.J.F.M. Kerkhof prof.dr. A. van Straten prof.dr. K.S. Bhui copromotor: dr. L.M. de Wit Leescommissie: prof.dr. Marcus Huibers prof.dr. Marrie Bekker prof.dr. Joop de Jong prof.dr. Karien Stronks dr. Derek de Beurs dr. Fatima El Fakiri table of contents Chapter 1 General Introduction 11 Chapter 2 Stigma for Common Mental Disorders Between Racial Minorities 27 and Majorities: A Systematic Review and Meta-analysis Chapter 3 Acculturation and Suicidal Ideation Among Turkish Migrants in 63 the Netherlands Chapter 4 Canına Kıymak “Crushing Life Energy”: A Qualitative Study 87 on Lay and Professional Understandings of Suicide and Help- seeking for Suicide among Turkish Migrants in the UK and in the Netherlands Chapter 5 Attempted Suicide and Suicide of Young Turkish Women in Turkey 115 and in Europe. A systematic Literature Review of Characteristics and Precipitating Factors Chapter 6 Protocol: Reducing Suicidal Ideation Among Turksih Migrants 175 in the Netherlands and in the UK: Effectiveness of an Online Intervention Chapter 7 Reducing Suicidal Ideation among Turkish Migrants in the 197 Netherlands and in the UK:A Feasibility and Pilot RCT of an Online Intervention Chapter 8 General Discussion 235 Chapter 9 Summary 253 Samenvatting (Summary in Dutch) 259 Özet (Summary in Turkish) 265 About the author 271 Publications 275 Acknowledgements 279 References 287 Appendices 313

¶”I had melancholy thoughts….. A strangeness in my mind, A feeling that I was not for that hour, Not for that place.” William Wordsworth, The Prelude

1

general introdUction 12 ― Chapter 1

rationale and aims

uicide is a global public health problem with serious consequences at individual and societal levels (Bertolote & Fleischmann, 2005). The international lifetime prevalence of suicidal ideation, plans and attempts in the general population in both high income and low income countries is S9.2%, 3.1% and 2.7%, respectively (Nock et al., 2008). One of the long-standing patterns in suicide epidemiology is the ethnic variation in rates and characteristics of suicidal behaviours (Haigh, Kapur & Cooper, 2015; Bhui, McKenzie & Rasul, 2007). There is also evidence for further variation particularly when stratified by region (whether urban or rural area), age and gender (Cooper et al., 2010). Suicide risks in ethnic minorities are elevated 2-4 fold when compared to their White counterparts in the UK (Bhui, Halvorsrud & Nazroo, 2018). In Europe, there is an increased risk for suicidal behaviours among Turkish migrant populations compared with the majority populations of the host countries (Razum & Zeeb, 2004; Garssen et al., 2006; van Bergen et al., 2019). The suicide attempt risk is increased 2-5 fold among a sub-group of Turkish migrant women aged between 14 and 25 when compared with the same aged women from the majority populations of the Netherlands (Burger et al., 2006), Germany (Lizardi et al., 2006) and Switzerland (Brückner et al., 2011) respectively. These differences in suicide epidemiology point to the patterns and circumstances leading to suicidal behaviours amongst some ethnic groups which might be different to the better known circumstances and factors identified for majority populations such as history of suicide attempts (Heredia-Montesinos, 2015). It is a well-documented phenomenon that suicidal people do not receive adequate psychological treatment. A worldwide survey on the treatment of suicidal people estimated that 56% of people living in high income countries receive help compared with only 17% in low income countries. (Bruaffers et al., 2011). These findings are comparable to data on accessibility for depression treatment (44% in high income; 7% in low income countries; Thornicroft et al., 2017). Ethnic inequalities in access to mental health care General introduction ― 13 exist worldwide and continue to be a cause for concern (Bhui et al., 2018). For instance in the UK, Black (21%) and South Asian (18%) ethnic groups presenting with self-harm are less likely to receive treatment compared to White populations (33%; Cooper et al., 2010). Similarly, in the Netherlands, Turkish migrants` mental health needs are less likely to be met compared to their Dutch counterparts (Turkish: 24.6%; Dutch:9.3%) during the help seeking stage (Fassaert et al., 2009b) and this is highlighted as one of 1 the key reasons for dropping out of mental health services among Turkish migrants (Hilderink et al., 2009; O`Brien, Fahmi & Singh, 2009). Although the research from the Netherlands represent the use of psychological services in general, it is known that the underutilisation of mental health care for suicidal behaviours mirror the general trends of ethnic disparity (Sheehan et al., 2018). Thus, ethnic minorities have been identified as representing vulnerable groups and have recently been prioritised in national suicide prevention strategies in many countries (e.g. Department of Health, 2012; Han et al., 2014). A number of specific barriers in the low uptake of services have been identified such as unemployment, or transportation problems. There are also cultural barriers such as the stigma attached to seeking help, mistrust in the health care system of the host country and the cultural mismatch between the service-user and the provider (Goldston et al., 2008; Fassaert et al., 2010; Snowden & Yamada, 2005). In recent years, e-mental health has been introduced to treatment provision, as an addition, or alternative to the conventional format of face to face service delivery. Offering treatments online is less costly and also removes the transportation barrier from the help-seeking process (Arshad et al., 2019). It is relatively easy to offer treatments through personal computers, mobile phones or tablets depending on the users’ language preferences and cultural values (Eylem et al., 2015). There are several randomised controlled trials (RCTs) demonstrating the effectiveness of e-mental health in treating suicidal ideation amongst general populations (e.g. van Spijker et al., 2014; De Jaegere et al., 2019; Arshad et al., 2019). However, evidence for the effectiveness of e-mental health among ethnic minorities with suicidal behaviours is currently lacking (Mishara & Kerkhof, 2013; Caplan et al., 2018). Given the assumed 14 ― Chapter 1 advantages such as flexibility and cultural relevance, e-mental health is promising for engaging with ethnic minorities and facilitating greater access to appropriate mental health care (Eylem et al., 2015). Overall, this thesis aims to improve the availability and provision of psychological interventions in the treatment of suicidal ideation. The thesis mainly focuses on the Turkish migrant populations in Europe given the widely reported disparity between the elevated risk for suicidal behaviours and their engagement with the available services (Fassaert et al., 2009b; Ünlü İnce et al., 2014a). Further, this thesis uses an e-mental health intervention for suicidal ideation, which was developed and tested by van Spijker and colleagues for the general population in the Netherlands (van Spijker et al., 2010). Specifically, this research seeks to understand the feasibility and clinical usefulness of the culturally adapted version of this intervention in improving engagement with Turkish migrants presenting with suicidal ideation. In the current introductory chapter, definitions of the constructs used in this thesis will be clarified, background research will be outlined and the remaining chapters are presented. definitions Historically, there has been much debate regarding the definition and conceptualisation of suicide and suicidal behaviours in the literature. One approach suggests a continuum between suicidal thought and attempt (Joiner, 2005), whereas another categorises these behaviours based on the intention to kill oneself, thereby differentiates suicide attempt (i.e. behaviours with an apparent intent to kill oneself) from self-harm (i.e. suicide-related behaviours with no intent to kill oneself; O`Caroll et al., 1996). Several attempts have been made to define a uniform set of terms for suicidal behaviours (e.g. De Leo et al., 2006). However, no consensus has been reached up until now. For the purposes of this thesis, the umbrella term “suicidal behaviours” is used to refer to suicidal ideation, suicide attempt, self-harm and suicide. The construct of “suicidal ideation” is defined as “thoughts about actively ending ones’ life, however ambivalent they may be” (van Spijker, 2012, p.8). This definition General introduction ― 15 acknowledges individuals’ ambivalence about having thoughts about suicide. Regardless of their own attitudes about having these thoughts and their potential in acting on them, they are still actively thinking about suicide. The term “attempted suicide” is used to indicate “a non-habitual act with nonfatal outcome that the individual, expecting to, or taking the risk to die or inflict bodily harm, indicated and carried out with the purpose of bringing about wanted changes” (De Leo 1 et al., 2006, p.14). The term ‘self harm’ is considered as part of the suicidal continuum and is defined as self-poisoning or self-injury irrespective of the apparent purpose of the act (National Institute of Clinical Excellence, 2019). This is because using “intent” as a criteria to separate self-harm from the spectrum of suicidal behaviours is imprecise, vague and not easily quantifiable (De Leo et al., 2006). Furthermore, there is an overlap between self- harm and the rest of the suicidal behaviours in the spectrum irrespective of the intent (De Leo et al., 2006). The term “suicide” refers to “an act with fatal outcome, which the deceased, knowing or expecting a potentially fatal outcome, has initiated and carried out with the purpose of bringing about wanted changes” (De Leo et al., 2006, p.12). The concept of race-ethnicity was defined based on the minority or majority classifications of the country of the included studies in this thesis. Furthermore, “migrant” is used to denote “those who have moved to a new country” (Ünlü İnce, 2014c). The concepts “first generation” refers to individuals who were born in country of origin and are settled in the respective host countries, while “second generation” refers to those who are children of the first generation and were born in the host countries. The concept of “stigma” is used to refer to “a mark of shame, disgrace or disapproval which results in an individual being rejected, discriminated against and excluded from participating in a number of different areas of society” (World Health Organisation, 2001). There are various definitions of mental illness stigma. The Mental Illness Stigma Framework (MISF, Fox et al., 2018) is used in this thesis to understand how stigma is differentially experienced in accordance with the perspective of the general public 16 ― Chapter 1 who often attributes the stigma (i.e. stigmatiser) to those who have mental illnesses (i.e. stigmatised; Fox et al., 2018). Additionally, perceived stigma refers to the shared perceptions and attributions of stigma between those who stigmatise and those who are stigmatised. The MISF has been chosen since this framework identifies specific stigma outcomes capturing specific stigma mechanisms (Fox et al., 2018). Another fundamental construct in this thesis is “acculturation” and refers to the phenomena which results when groups of individuals from different cultures come into continuous first-hand contact, with subsequent changes in the original cultural patterns of either or both groups.” (Berry, 1997). This definition is based on Berry`s bi-dimensional framework suggesting that acculturation processes are essentially the same for all. Even though this universalist approach has been criticised for simplifying the construct (e.g. Castles, 2010; Borges et al., 2009), there is well-documented evidence supporting its cross-cultural validity (Berry, 2009). The term “cultural adaptation” is defined as “the systematic modification of an evidence-based treatment or intervention protocol to consider language, culture, and context in such a way that it is compatible with the individual’s cultural patterns, meanings, and values” (Bernal, et al., 1995). Lastly, “e-mental health” and “online intervention” are used interchangeably and refer to “psychological interventions delivered via websites and browser based applications” (Christensen, Batterham & O’Dea, 2014). ethnicity and mental illness stigma Stigma is a multidimensional construct with emotional and behavioural consequences for those who are affected such as social isolation and discrimination within social circles and/or within institutions (Nadeem et al., 2007; Lee et al., 2009). Stigma also prevents individuals from using available mental health services (Clement et al., 2015b; World Health Organisation, 2017). These consequences might be worse for ethnic minorities who often face other social adversities such as migration, poverty and other minority statuses (e.g. sexual minority) in their respective host countries (Bhui, 2010a; Bhui, 2019). General introduction ― 17

This is known as the intersectional impact of stigma, and is arguably accounts for the increased risk of mental illnesses and suicidal behaviours among some ethnic minority groups relative to others (Bhui et al., 2018). The supportive evidence for the intersectional impact of stigma mainly comes from cross-sectional studies (see Chapter 2). For instance, one large scale study from America showed that among women of low socioeconomic background with depression, those 1 from Black ethnic backgrounds reported more stigma related concerns about having depression compared to the women from White backgrounds (Nadeem et al., 2007). Similarly, in the case of Turkish migrants in Europe, the strong stigma attached to having suicidal thoughts (e.g. Eylem et al., 2016; Heredia Montesinos et al., 2018) often coincides with social adversities such as unemployment and personal incompetence attributions of mental illness. Arguably, these factors have collective impact on their distress, under-utilisation of mental heath services and ultimately on the increased risk for suicidal behaviours. Even though the preceding intersectionality argument is appealing, its impact for the risk of suicidal behaviours among particular ethnic groups is not conclusive. Stigma related to suicide has been ignored until very recently (Rimkeviciene et al., 2019) and whether suicide-related stigma is a separate construct remains unclear (Rimkeviciene et al., 2019). Arguably, stigma has a similar mechanism in different circumstances (Fox et al., 2018). For instance, both mental illness stigma and suicide-related stigma happen in the same socio-cultural context leading to the same consequences such as withdrawing from social life or not seeking help. Thus, even though the content of those constructs are different, furthering our understanding of mental illness stigma for instance, is likely to have an implication on other types of stigma. In recent years, there is growing literature of cross-sectional studies comparing the differences between ethnic minorities and majorities in various mental illness stigma outcomes (Lee et al., 2009; Makowski & von dem Knesebeck, 2017). Thus, a comprehensive up-to-date meta- analysis, summarising the evidence base on racial and/or ethnic differences in mental illness stigma is necessary. It also has important policy implications for the global anti- 18 ― Chapter 1 stigma strategies with the intention to improve the provision of mental health services for culturally diverse populations. The following section will focus on Turkish migrant populations in Europe since they represent one of the most vulnerable groups for suicidal behaviours (Eylem et al., 2016; Heredia Montesinos et al., 2018; van Bergen et al., 2019). tUrkish migrant popUlations in eUrope Turkish migrant populations are one of the largest ethnic minority populations residing in Europe. Some of the largest Turkish migrant populations live in Germany (13.52%) and in the Netherlands (2.3%; Kilberg, 2014; Statistisches Bundesamt, 2019). In the UK, the Turkish populations are the 15th biggest minority group (D`angelo, Garip and Kaye, 2013). They are mainly concentrated in 3 boroughs of London: Enfield, Hackney and Harringay and constitute 7.5% of the population in these boroughs (D`angelo et al., 2013). Their migration status often coincides with social adversities such as socio-economic disadvantages in the preceding host countries. For instance, in the Netherlands the unemployment rate is 2 times higher among Turkish migrants than among their Dutch counterparts (Centraal Bureau Statistics, 2016). Similarly in the UK, the unemployment rate among Turkish migrants was estimated to be 7% (against 4% for the general population; D`anglo et al., 2013). Turkish populations migrated to Europe for different historical, political and economic reasons (Enneli et al., 2005). The history of the Turkish population in Germany and in the Netherlands dates back to 1961 when a convention agreement was signed, therefore the reasons for migration were mostly economic. With respect to the UK, the outflow of migration from Turkey happened after mid-seventies and peaked during nineties mainly due to political instability in Turkey (Enneli et al., 2005). Migration to Europe was often from the rural areas in the beginning. However, more recent migration was from urban areas (Enneli et al., 2005). There has also been an increase in the number of political refugees from Turkish origin in Europe in recent years (Özdemir et al., 2019). Despite the diversity, there are commonalities shared by all as a result of their history of General introduction ― 19 interaction. For instance, traditionalism (i.e. a strong religious and ethnic identification with the culture of origin) is common especially amongst the Turkish migrants in Germany, Belgium and the Netherlands, emphasising conformity and the importance of family “honour” (Güngör, 2008). The most recent group of Turkish migrants might diverge from these traditional values as they are often highly educated, less religious and have more liberal political ideologies (Özdemir et al., 2019). 1 Turkish migrants in Europe present cross-national consistency in prevalence rates and characteristics of suicidal behaviours (see Chapter 5, Chapter 7). The most common explanations for the elevated risk of suicidal behaviours amongst particular ethnic groups relative to the others focus on migration related factors such as acculturation difficulties in the host countries (Lester, 2013). Since migration related factors are relevant to all migrants, perhaps they are not sufficient to fully explain the differences in suicide epidemiology between Turkish migrants and the other minority groups (see Eylem et al., 2019). Thus in the following section, the role of acculturation and other possible contributory factors such as cultural meaning and cultural continuity are considered in explaining the risk for suicidal behaviours. social and cUltUral risk factors

Acculturation Acculturation is a significant aspect of the migration process (Berry, 1997). Acculturation is viewed as an inevitable process people undergo in an effort to manage and cope with stressors and changes brought upon by migration and by being in a prolonged contact with a new, host culture (Berry, 1997). Consequently, there exists an interwoven relationship between coping and acculturation for individuals undergoing cultural change and transition (Bhui, 2019). According to the Berry`s bi-dimensional framework, individuals either have more tendency to maintain their link with their culture of origin (i.e. cultural maintenance), or they have more tendency to participate in the social life of the host country (i.e. participation; Berry, 1997) in the acculturation process. Culture 20 ― Chapter 1 conflict occurs when there is a discrepancy between expectations (assumptions) and actuality (experience) between the individual, the members of the host country and the members of the culture of origin (Bhugra, 2004). Very often, this conflict is linked with psychological distress, mental health problems and suicidal behaviours especially if the expectations and the actuality of the culture of origin are very different than those of the host culture (Bhugra, 2004; Lester, 2013). Support for the difficulties in acculturation processes among Turkish migrants comes from cross-sectional research in Belgium (Phalet & Schonpflug, 2001), Germany (Aichberger et al., 2015) and the Netherlands (Fassaert et al., 2010). Turkish migrants with greater orientation for cultural maintenance often report more psychological distress in their contact with other minority and majority populations especially when they feel their ethnic identity is incompatible with the majority culture (Aichberger et al., 2015; Fassaert et al., 2010; Phalet & Schonpflug, 2001; Güngör & Bornstein, 2009). Although acculturation difficulties are well-documented as risk factors (Bhugra, 2004; Lester, 2013), currently little is known about how acculturation generates a potential risk for suicidal behaviours among some ethnic groups relative to the others. There is therefore a need for theoretically informed models to further the current understanding of the circumstances which enhance or diminish (i.e. moderators) the risk of suicidal behaviours, as well as identify the specific mechanisms of how the risk for suicidal behaviours is generated (i.e. mediator; Kleiman & Anastis, 2015).

Cultural meaning of suicide One of the alternative explanations for the ethnic variation in suicide epidemiology is the cultural meaning attributed to suicide. It is argued that holding cultural and religious beliefs, which would sanction or honour suicide, may have more relevance to suicide risk than other well-known risk factors such as the presence of mental illness (Hjelmeland, 2011; Colucci & San Too, 2015; Lester, 2013). This phenomenon is named as the “pathogenic effect of culture” meaning that cultural meaning of suicide contributes to General introduction ― 21 the distress that people may suffer and the distress may then contribute to the occurrence of suicidal behaviours (Colucci, 2013). The qualitative studies point to the gender and culture specific characteristics of suicidal behaviours among Turkish-speaking populations (e.g. Heredia Montesinos et al., 2018; Cetin, 2015; 2017). For instance in Germany, focus group interviews with Turkish migrants highlight the fear of dishonouring themselves and/or their family as a key 1 reason for not seeking help because of suicidal behaviours (Heredia Montesinos et al., 2018). The main methodological consideration of the existing qualitative studies is that cross-cultural research is scarce. This limits our understanding of the differences in the meanings attributed to suicidal behaviours in relation to the differences in the socio- cultural contexts (of the host countries; Colucci, 2013).

Cultural continuity The cultural continuity phenomenon is known as the tendency to import the characteristics and risk factors for suicidal behaviours from their country of origin into the countries they migrate to (Lester, 2013). The consistency in suicide epidemiology between migrant communities and their counterparts from the country of origin is often explained with this hypothesis (Lester, 2013) Support for the cultural continuity theory comes from the epidemiological studies in Turkey (Bağlı & Sever, 2003) and in several European countries such as Germany (Brückner et al., 2011; Heredia Montesinos, 2015), the Netherlands (Burger et al., 2006) and Switzerland (Yilmaz & Riecher-Rossler, 2008). Results draw attention to women of Turkish descent, aged between 14 and 25 demonstrating disproportionate rates of suicidal behaviours in the host countries and in their country of origin. Arguably, there is continuity in the traditional patriarchal family structure between Turkey and Europe, restricting the life choices of women in the host countries such as not being allowed to go out without a companion or male relative (van Bergen et al., 2010). In this context, suicide might seem like the only way to escape from this coercive pattern and the associated feelings (e.g. van Bergen et al., 2010; Schouler-Ocak 2015; Rezaeian, 2010; Canetto, 2015). 22 ― Chapter 1

The cultural continuity argument is well documented, but it remains unclear to what extent there is continuity (or change) in the social and cultural factors contributing to suicidal behaviour among this sub-group of Turkish women. The main reason is that none of the studies compare the migrant population with their counterparts from the country of origin (Colucci, 2013). In addition to these risk factors, there is an unequal access to the available services among migrants and ethnic minorities (Bhui et al., 2018). In the final part of this chapter, the available treatments for suicidal behaviours are outlined and the possible benefits of recent technological advancements, such as e-mental health for ethnic minorities are discussed. available treatments for sUicidal behavioUrs There is evidence to suggest that psychological and social interventions may be effective in treating suicidal behaviours (Hawton et al., 2017; Hawton et al., 2015; Hetrick, Robinson, Spittal and Carter, 2016; Turner, Austin & Chapman, 2014). There is also preliminary support for the efficacy of a number of interventions in managing suicidal behaviours such as Cognitive Behavioural Therapy (CBT), Dialectical Behavioural Therapy (DBT; Hawton et al., 2017; Hawton et al., 2015; Hetrick et al., 2016; Turner et al., 2014), and Mentalization-Based Treatment (MBT; Bateman & Fonagy, 1999; Rossouw and Fonagy, 2012; Bateman & Fonagy, 2008). CBT and DBT are based on the general cognitive model, suggesting that the way someone interprets a situation determines their emotional and behavioural reactions to those situations (Beck, 1995; Wenzel et al., 2009). These so-called interpretations are distorted by errors in thinking such as overgeneralisation and emotional reasoning. Thus, interventions, which are based on the cognitive model, aim to restructure thinking errors so that individuals can make sense of situations from a more realistic point of view (Beck, 1995). Mentalization is the capacity to understand actions in terms of thoughts and feelings (Rossouw & Fonagy, 2012). When mentalizing is compromised in interpersonal relationships, negative General introduction ― 23 thoughts are experienced in greater intensity leading to an urgent need for distraction. In this context, suicidal behaviours may serve as distraction (Rossouw & Fonagy, 2012). There are important limitations with the preceding interventions. They are typically costly in terms of resources, the necessity of trained therapists, and the duration of therapy typically required (Arshad et al., 2019). Further, the majority of these interventions rely on face-to-face contact (delivered at a group, family, community or 1 individual level; Arshad et al., 2019). These interventions may therefore not be accessible for many individuals, due to geographical (e.g. living in rural settings with limited mental health resources, travel distance to appointments), social (e.g. barriers related to stigma), organisational (e.g. waiting times and service availability) or even financial reasons (e.g. in healthcare contexts where clients must pay for their treatment; Department of Health, 2017; Institute of Medicine, 2002; Leigh & Flatt, 2015; Poppleton & Gire, 2017). e-mental health In recent years, e-mental health interventions and mobile health technologies have been suggested as viable solutions to overcome the preceding barriers restricting the utilisation of available psychological therapies (Cuijpers et al., 2019; Harper-Shehadeh et al., 2016). Recent meta-analytic studies suggest that the treatment of suicidal ideation might also benefit from e-mental health interventions (Arshad et al., 2019). Results indicate a significant beneficial treatment effect on reducing suicidal ideation g = -0.26 (95% CI: -0.48, -0.03; Arshad et al., 2019). Van Spijker and colleagues (Kerkhof & van Spijker, 2011) developed an e-mental health intervention for treating suicidal ideation based on the cognitive model (Beck, 2005). There are RCTs showing the effectiveness of this intervention in reducing suicidal ideation compared to the treatment as usual (i.e. information website) in general Dutch (d=0.2; van Spijker et al., 2014) and Belgian populations (d= 0.34; De Jaegere et al., 2019), but it showed no effect when implemented in an Australian population (van Spijker, et al., 2018). The contradictory findings may highlight the variations in attitudinal (e.g. mental illness stigma), geopolitical (e.g. accessibility to internet) or cultural barriers between 24 ― Chapter 1 and within countries. Thus, it is important to consider the differences in socio-cultural contexts whilst implementing e-health interventions. One way of optimising e-mental health interventions is to tailor the content based on the characteristics of the target populations. There is considerable evidence suggesting that adapting e-mental health interventions according to the cultural context of the users can increase their effectiveness and efficacy (Harper-Shehadeh et al., 2016). Overall, culturally adapted e-mental health interventions might make psychological help more accessible to ethnic minorities by removing logistical barriers such as distance or transport issues (Cuijpers et al., 2019). They might also remove stigma and shame associated with help-seeking (Rathod et al., 2019). Thus, culturally adapted online interventions seem promising for ethnic minorities in general (Harper-Shehadeh et al., 2016) and for Turkish migrants specifically (Ünlü İnce et al., 2013). To date, whether this form of service provision improves engagement with Turkish migrants in treatment of suicidal behaviours has not been investigated. oUtline of the thesis To address the current knowledge gap on the issues outlined above, the present thesis employs various qualitative and quantitative methodologies. Chapter 2 includes a systematic review and meta-analysis examining the differences in mental illness stigma for common mental disorders (CMDs) between racial minorities and majorities. The chapter also investigates whether the quality of the studies and racial classification are related with the effect size. Chapter 3 investigates the relationship between acculturation and suicidal ideation among Turkish migrants in the Netherlands. In this chapter the focus is placed on furthering the current knowledge on the theoretically informed models for suicidal behaviours among Turkish migrants, including mediators and moderators. Chapter 4 investigates the relevance of cultural conceptualisations of suicide and help- seeking for suicide in Turkish migrants through a cross-cultural qualitative study in the Netherlands and in the UK. Furthermore, in an effort to investigate the relevance of the cultural continuity thesis, Chapter 5 employs a systematic literature review General introduction ― 25 comparing and synthesising the empirical evidence of the characteristics and precipitating factors of suicide and attempted suicide of Turkish women in Europe and in Turkey. Chapter 6 describes the protocol for the RCT of an e-mental health intervention for suicidal ideation, which was culturally adapted according to the Turkish migrants in the Netherlands and in the UK. Chapter 7 presents the results of the feasibility and a pilot RCT comparing the culturally adapted e-mental health intervention with treatment as 1 usual (i.e. information website). Finally, Chapter 8 concludes with a general discussion of the key findings, which also includes limitations, suggestions for future research and implications for clinical and research practice.

2

STIGMA FOR COMMON MENTAL DISORDERS IN RACIAL MINORITIES AND MAJORITIES: A SYSTEMATIC REVIEW AND METAANALYSIS

pUblished as Eylem, O., de Wit, L., van Straten, A. et al. (2020) Stigma for common mental disorders in racial minorities and majorities a systematic review and meta-analysis. BMC Public Health 20, 879 . https://doi.org/10.1186/s12889-020-08964-3 28 ― Chapter 2

abstract

Background There is a strong stigma attached to mental disorders preventing those affected from getting psychological help. The consequences of stigma are worse for racial and/or ethnic minorities compared to racial and/or ethnic majorities since the former often experience other social adversities such as poverty and discrimination within policies and institutions. This is the first systematic review and meta-analysis summarizing the evidence on the impact of differences in mental illness stigma between racial minorities and majorities.

Methods This systematic review and meta-analysis included cross-sectional studies comparing mental illness stigma between racial minorities and majorities. Systematic searches were conducted in the bibliographic databases of PubMed, PsycINFO and EMBASE until 20th December 2018. Outcomes were extracted from published reports, and meta- analyses, and meta-regression analyses were conducted in CMA software.

Results After screening 2,787 abstracts, 29 studies with 193,418 participants (N=35,836 in racial minorities) were eligible for analyses. Racial minorities showed more stigma than racial majorities (g=0.20 (95% CI: 0.12~0.27) for common mental disorders. Sensitivity analyses showed robustness of these results. Multivariate meta-regression analyses pointed to the possible moderating role of the number of studies with high risk of bias on the effect size. Racial minorities have more stigma for common mental disorders when compared with majorities. Limitations included moderate to high risk of bias, high heterogeneity, few studies in most comparisons, and the use of non-standardized outcome measures. Stigma for Common Mental Disorders ― 29

Conclusions Mental illness stigma is higher among ethnic minorities than majorities. An important clinical implication of these findings would be to tailor anti-stigma strategies related with mental illnesses according to specific racial and/or ethnic backgrounds with the intention to improve mental health outreach.

2 30 ― Chapter 2

introdUction

ommon mental disorders (CMDs) such as depression and anxiety disorders are highly prevalent, disabling and costly with diminished quality of life, medical morbidity and mortality (Smith et al, 2007; Üstün et al., 2004; Cuijpers et al., 2009; Cuijpers et al., 2019). It is estimated that Cevery year almost one in five people among the general population worldwide suffers from common mental disorders (CDMs), such as depression and anxiety (Kessler et al., 2013; Steel et al., 2014). Even though so many people are affected by CMDs globally, there is a strong stigma attached to CMDs and those who have them (World Health Organisation, 2017). Mental illness stigma is a multidimensional problem causing great burden on those who are affected (World Health Organisation, 2017; Korszun et al., 2012). Not only does it determine negative public opinion and discrimination against people with mental illnesses (Korszun et al., 2012; Nadeem et al., 2007; Crenshaw, 1989) but it also leads to not to seek or adequately participate in psychological treatment (Nadeem et al., 2007; Lee et al., 2009; Clement et al., 2015). There are various definitions of mental illness stigma in the current stigma literature. Recently, the Mental Illness Stigma Framework (MISF) has been proposed (Fox et al., 2018) to define the different types of stigmas and mechanisms associated with them. Thus, how stigma is experienced differs depending on the perspective of general public who often attributes stigmas (i.e. stigmatiser) to those who have mental illnesses (i.e. stigmatised) (Fox et al., 2018). There are also shared perceptions and attributions between stigmatiser and stigmatised (i.e. perceived stigma). According to this model, there are different cognitive, affective and behavioural mechanisms associated with each perspective. The cognitive mechanisms are stereotypes referring to the collectively agreed upon negative beliefs about an individual with a mental illness (e.g. dangerousness, weakness). The affective mechanisms are prejudices which are emotional reactions generated by stereotypes such as, fear, anger and pity. The behavioural mechanisms are named as discrimination such as, withholding help, avoidance, segregation or coercion Stigma for Common Mental Disorders ― 31

(Corrigan & Watson, 2002; Fox et al., 2018). The impact of stigma on an individual`s life can be understood in terms of three components: 1) Experienced stigma, referring to the day-to-day experiences of stereotypes, prejudice and discrimination from others, 2) anticipated stigma, the expectation to be a target of a stereotype, prejudice or discrimination, and 3) internalized stigma, which is the application of mental illness stigma to oneself such as believing that they are dangerous to others or they are incompetent (Fox et al., 2018). There are differences in the extent of the impact of mental illness stigma depending 2 on the racial and/or ethnic background of those who are affected (Fox et al., 2018). Early research on the influence of ethnicity on the mental illness stigma indicated that compared to the White group, the non-White group perceived someone with mental illness as more dangerous (Corrigan et al., 2007) and expressed greater need for segregation than the White group (Corrigan et al., 2007). These results were replicated by more recent research comparing Asian Americans (Cheng et al., 2015; Fogel et al., 2005; Georg Hsu et al., 2008), African Americans (Anglin et al., 2006; Brown et al., 2010; Coner et al., 2009) and Hispanics (Givens et al., 2007; Caplan et al., 2020) with European Americans (White). Further, the variation in mental illness stigma could be even more so among specific ethnic groups within broad racial categories (Subramaniam et al., 2017). For instance, in a large scale study representative of the ethnic groups in Singapore, Subramaniam and colleagues found that, those of Indian ethnicity who also had low socio-economic status, perceived individuals with mental illness as more dangerous and unpredictable, and desired more social distance compared to those of Malay and Chinese backgrounds (Subramaniam et al., 2017). Thus, there is intersectionality in experiences of stigma (Makowski et al., 2017; Fox et al., 2018). This perspective emphasises that the consequences of stigma are worse for some racial and/or ethnic groups who have for instance, personal incompetence attributions of mental illness (Fox et al., 2018) and who also face with other forms of “minority stress” and adversities such as interpersonal and structural discrimination within policies and institutions and low socio-economic background (Nadeem et al., 2007; Lee et al., 2009; Clement et al., 2015). 32 ― Chapter 2

To date, there are no prior meta-analysis investigating racial and/or ethnic differences in mental illness stigma. This could be due to the inconsistency in how stigma mechanisms have been defined and measured (Fox et al., 2018). One implication of this gap in the literature is the lack of information about the evidence base for developing culturally-relevant anti-stigma interventions (Gronholm et al., 2019). Since CMDs are highly prevalent globally, it is important that psycho-social interventions focus on changing the negative stereotypes (e.g. I am incompetent) and/or discriminatory behaviours (e.g. social withdrawal) related with CMDs among racial and ethnic minorities specifically (Gronholm et al., 2019; Corrigan & Watson, 2002). Nevertheless, there are promising theoretical developments in the stigma field such as the MISF. There is also growing number of studies examining the ethnic variations in stigma for CMDs in Western as well as in Non-Western countries. Thus, the objectives of the current study are twofold. We examine the differences in mental illness stigma between racial minorities and majorities. We expect that racial minorities have more mental illness stigma for CMDs compared to majorities. We also investigate if there are variability in mental illness stigma between racial minorities depending on race, quality of the studies and types of stigma outcomes (self-report vs vignette). methods

Identification and selection of studies A review protocol was developed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA)-statement (www.prisma-statement.org) (CONSORT, 2018). The protocol for this meta-analysis was registered at PROSPERO (CRD42018091080).

Inclusion criteria The searches were limited to the following criteria: 1) Peer-reviewed papers, 2) Racial minorities (i.e. defined based on the classification of the country of the included studies) Stigma for Common Mental Disorders ― 33

2) Racial majorities (i.e. defined based on the classification of the country of the included studies) 3) Adults aged 18 and above and 4) Participants with/without common mental disorders (i.e. common mental disorders are identified as depression and anxiety spectrum disorders), 2) Empirical studies with cross-sectional designs measuring mental illness stigma about common mental disorders among racial minorities in comparison to majorities 3) Studies were not limited to the European populations only and studies carried out in other continents were included. 2 Exclusion criteria The exclusion criteria: 1) Publications focusing on stigma about help-seeking, HIV, physical disorders or sexual minorities (if they are not from a racial minority group), 2) Publications focusing on stigma about severe mental health disorders (e.g. schizophrenia), 3) Empirical studies without a comparison group (studies which are not comparing different racial groups were excluded), 4) Qualitative studies and 5) Publications focusing on adolescent and children sample.

Quality assessment We assessed the quality of the included studies using the Effective Public Health Practice Project Quality Assessment Tool (EPHPP). This tool assesses possible sources of bias in observational studies and RCTs. Since we have included cross-sectional studies, the following domains of the tool were used in this study: (1) selection bias; (2) study design; (3) confounders; (4) data collection method and; (5) analyses. The studies received an overall assessment in one of the following: 1) high risk of bias (studies which scored high risk of bias in 3 or more of the assessment domains); 2) moderate risk of bias (studies which scored high risk of bias in 2 of the assessment domains) and; 3) low risk of bias (studies which scored high risk of bias in 1 of the assessment domains). Assessment was carried out by two independent assessors and disagreements were solved through discussions. 34 ― Chapter 2

Data extraction A customised data extraction form was generated and included the following characteristics (method of recruitment into the study such as, through community, clinical samples or other recruitment type), target group (adults in general, older adults, student population or other target group), types of stigma perspectives, types of stigma mechanisms and types of outcome measures used to measure stigmas (self-report instruments, vignettes). Vignettes are case descriptions of an individual, presenting symptoms of a CMD (Subramaniam et al., 2017). Covidence, online software for screening and data extraction for systematic reviews and meta-analyses, was used for the review and extracting data. First, the results of the online database searches were imported to covidence. Two reviewers had personal accounts and selected papers independently in a random order. A third reviewer carried out the reference list search and selected papers. The titles of all studies were screened, and the abstracts of the studies were checked regarding the inclusion criteria. When no definitive decision could be made based on the abstract, the original papers were used. Discrepancies between the reviewers’ selections were resolved through discussions. If not resolved, the opinion of a fourth researcher was sought. The corresponding author filled in the extraction form.

Conceptual frameworks We conceptualised mental illness stigma based on the MISF (Fox et al., 2018). The types of perspectives, measured by the studies, were categorised into three groups: 1) the perspective of the stigmatiser (i.e. Public attitudes and beliefs that other people devalue or discriminate against individuals with mental illness. The specific components are: stereotype, prejudice and discrimination); 2) The perspective of the stigmatised (i.e. Personal beliefs, attitudes and perceived anger for having a mental illness. The specific components are: experienced stigma, anticipated stigma and internalized stigma) and; 3) The perceived stigma (i.e. shared experiences of stereotypes, prejudices and discrimination between people who stigmatise and who are stigmatised). Stigma for Common Mental Disorders ― 35

The concept of race-ethnicity was defined based on the minority or majority classifications of the country of the included studies. Because there were many ethnic groups (e.g. Chinese, Indian) within the included studies, we decided to use broad racial categories (e.g. Asian, Black) in order to make the studies comparable (Conelly, Gayle & Lambert, 2016). Six racial groups were identified in consultation with the categories previously defined by Ünlü İnce and colleagues (Ünlü İnce et al., 2014c). These categories were: Black (African background), Asian, Hispanic (Latin American and Spanish background), Native American (referring to the indigenous people of North America), 2 and White (Caucasian and white European) and other (people from racial-ethnic minority group who could not be identified in one of these categories).

Outcome For each comparison between ethnic groups in stigma, the effect size indicating the difference between groups was calculated (Hedge`s g). Effect size of 0.8 was accepted as large, effect size of 0.5 was accepted as moderate and effect size of 0.2 was accepted as small (Cohen, 1988). Effect sizes were calculated by subtracting the means of stigma between ethnic minorities and majorities and dividing the result by the pooled standard deviation. If means and standard deviations were not reported, we used the procedures of the Comprehensive Meta-Analysis software (see below) to calculate the effect size using dichotomous outcomes; and if these were not available either, we used other statistics (such as t-value or p-value) to calculate the effect size. In order to calculate effect sizes we used all the outcomes examining stigma [such as Perceived Devaluation and Discrimination Scale (PDD) (Link, 1987), Internalized Stigma of Mental Illness Scale (ISMI) (Ritsher et al., 2003), Community Attitudes to Mental Illness scale (CAMI) (Taylor and Dear-see Aznor-Lou et al., 2016); see table 1 for the outcome measures]. The decision on which outcome measure is capturing which specific stigma perspective and mechanism was based on Fox and colleagues` classification (Fox et al., 2018). 36 ― Chapter 2

Analyses To calculate pooled mean effect sizes, we used the computer programme Comprehensive Meta-Analysis (version 3.3070; CMA). Because we expected considerable heterogeneity among the studies, we employed a random effects pooling model in all analyses. As a test of homogeneity of effect sizes, we calculated the I2 statistic, which is an indicator of heterogeneity in percentages. A value of 0% indicates no observed heterogeneity, and larger values indicate increasing heterogeneity, with 25% as low, 50% as moderate and 75% as high heterogeneity (Higgins, Thompson, Deeks & Altman, 2003). We calculated 95% confidence intervals around I2 (Ioannidis, Patsopolous & Evangelou, 2007) using the non-central chi-squared based approach within the heterogi module for Stata (Orsini et al., 2006). We tested publication bias by inspecting the funnel plot on primary outcome measures and by Duval and Tweedie`s trim and fill procedure (Duval & Tweedie, 2000), which yields an estimated effect size after publication bias has been taken into account (as implemented in CMA). We also conducted Egger`s test of the intercept to quantify the bias captured by the funnel plot and to test whether it was significant. We also examined whether specific characteristics of the studies were related to the effect sizes. We conducted subgroup analyses according to the mixed effects model, in which studies within subgroups are pooled with the random effects model, while the tests for significant differences between subgroups are conducted with the fixed effects model. The priori decided sub-groups were: ethnicity, type of stigma outcome, type of stigma perspective and the quality of the studies. Further, we used multi-variate meta- regression analyses as implemented in CMA. resUlts

Selection and inclusion of the studies After examining a total of 2,787 abstracts (1,806 after removal of duplicates), we retrieved 1,806 full text papers for further consideration. We excluded 1,732 of the retrieved papers. Stigma for Common Mental Disorders ― 37

The PRISMA flowchart describing the inclusion process, including the reasons for exclusion, is presented in Figure 1. A total of 29 studies were included in the quantitative synthesis.

Figure 1. PRISMA flow chart of the studies

2 38 ― Chapter 2

Characteristics of the included studies Selected characteristics of the included studies are presented in Table 1. Of all the studies included in the analysis, (N=18) 62% were conducted in the United States and (N=3) 10% were conducted in the United Kingdom. The rest of the countries were: Singapore (N=2) 6%; Australia (N=2) 6%; Spain (N=1) 3%; Nigeria (N=1) 3%; Germany (N=1) 3%; Canada (N=1) 3% and; Korea (N=1) 3%. Regarding the participant characteristics, a total of 157,582 (81%) participants were from the majority racial groups and a total of 35,836 (19%) were from the minority racial groups. Minority and majority statuses were defined based on the country of the included study. Of all the participants from the racial majority groups, 96.06% (N=151,383) were White, 2.69% (N=4,239) were Asian, 0.06% (N=91) were Hispanic and 1.18% (N=1,869) were Black. Further, of all the participants from the racial minority groups, 11.30% (N=4,051) were Hispanic, 57.09% (N=20,462) were Black, 23.11% (N=8,285) were Asian, 9.31% (N=3,338) were from other racial background. With respect to the condition of the CMD, more than half of the studies (N=17) investigated various CMDs and/or not specified the type of CMD in the study, whereas (N=12) 41% studies investigated depression. A considerable number of the studies (48%) recruited community sample (N=14), 24% recruited student sample (N=7), 14% recruited other sample (e.g. geriatric population) (N=4) and 14% recruited participants from clinical populations (N=4). Regarding the outcomes, (N=7) 24% used a vignette approach to measure stigmas whereas, (N=22) 76% used self-report questionnaires. Stigma for Common Mental Disorders ― 39

Results 2 Asian majorities (M=35.64) had higher perceivedAsian stigma majorities than (M=35.64) Asian minorities (M=37) (p=0.007) Asian (M=20.9), Other (M=23.0) Black (M=23.5) and White Black (M=23.5) Other (M=23.0) Asian (M=20.9), had less stereotypes attitudesminority less (M=23.9) favourable (i.e. in authoritarianism) against those with compared CMDs to White majority (M=25.0) groups had 26.9) and Black OtherAsian (M=27.0) (M= (M=25.4) less attitudes stereotypes compared in to benevolence) favourable (i.e. and majority (M=27.7) minoritywhite (M=27.8) had less stereotypes Black (M=36.4) OtherAsian (M=36.2) (M=34.3) attitudes toward supporting favourable those(i.e. with CMDs) and majority (M=37.5) compared to White minority (M=37.3) had discrimination more least favourable (i.e. Asian (M=13.7) group attitudes towards compared those to Other with (M=15.3) CMDs) majority (M=16.5) White minority (M=16.6) Black (M=15.5) There was stereotype more perceived dangerousness) (i.e. against with people depression (p<.05) among Black compared group to White (t=2.14) group There was tendency less prejudice against (i.e. to blame) with people depression (p<.05) among Black compared group to White (t=-2.33) group punishment) endorse to tendency (i.e. discrimination less was There against with people depression among Black compared group to (p<.001) White (t=-3.91) group There were no ethnic differences in discrimination against people with with people against discrimination in differences ethnic no were There mental illness (p=0.14)

• • • • • • • • • Self-report PDD Self-report CAMI-23 Vignette Vignette Study- constructed Self-report SDS Outcomes Other Community Community Community Recruitment 3055)

=56) =913) =92) =82) =545) =7) = =118) =58) =1869) maj min min min maj min min min min maj =1668) maj Asian (N Black (N Other (N Other Black (N Asian (N White (N Black (N White (N Other (N Other Racial groups White (N Asian (N Type of CMD of Type Various Various Various Various Country Nigeria USA Catalonia, Spain North Korea Selected characteristics and main findings the of cross-sectional studies comparing ethnic minorities with majorities stigma on = 29) outcomes (N

Study Adewuya, 2008 Anglin 2006 Aznar-Lou Aznar-Lou 2016 Ahn 2015 Table 1. 40 ― Chapter 2 Results There was discrimination(i.e. more desire social for distance) against a person SD=1.06) with depression among Asian (M=3.16, group p=0.004 SD=1.18) compared to White (M=2.80, There was discrimination more less willingness (i.e. to hire and rent) against a person with depression among Asian (M=5.54, group p=0.008 compared SD=1.82) to White (M=6.00, SD=1.64) There against blame) was a person prejudice more (i.e. with depression compared to White (M=3.73, among SD=1.63) Asian (M=4.18, group p=0.02 SD=1.78) There against anger) was prejudice more a person (i.e. with depression compared to White (M=2.58, SD=1.90) among Asian (M=3.26, group p=0.002 SD=1.72) There against fear was someone with prejudice more someone) of (i.e. compared to SD=1.96) depression among Asian (M=3.78, group p=0.54 SD=2.01) White (M=3.32, There were and differences no WhiteSD=4.1) betweenBlack (M=31.3, groups SD=4.8) in perceived stigma (p=.55) (M=31.0, There were differences no and White betweenSD=11) Black (M=65.9, groups in internalised SD=11.4) (M=65.0, stigma (p=.42) There was higher anticipated stigma among minority Hispanic group compared to the majority (p=0,015)

• • • • • • • • Vignette AQ Self-report ISMI PDD Study- constructed Outcomes Community Community Clinical Self-report Recruitment =229) =206) =231) =220) maj maj min min =86) =91) maj min Black (N Asian (N Hispanic (N White (N Racial groups Hispanic (N White (N Type of CMD of Type Various Depression Depression Country USA USA USA Continued.

Study Brown 2010 Caplan 2011 Cheng 2015 Table 1. Stigma for Common Mental Disorders ― 41

Results 2 There was perceived more stigma perceived attitudes (i.e. others of compared SD=7.58) depression)about among Other (M=17.82, group SD=5.36) F=32.95 to White (M=9.03, There was stereotype more personal (i.e. attitudes toward depression) compared SD=6.13) depressionabout among Other (M=25.16, group F=10.78 SD=8.79) to White (M=19.35, There was differences no in perceivedstigma between 2.61, Black (M= groups .29) t [246]=-0.58 SD= 2.59, 0.28) andSD= White (M= There was internalised more SD=0.30) stigma among Black (M=2.18, .035). p= [246]=-2.118, (t group SD=0.30) compared to White (M=2.10, There was perceived more stigma among Black = 2.90, (M = 0.75) SD < .001) (p group compared to White = 2.32, (M = 0.55) SD There was internalised more stigma among = SD Black = 2.75, (M compared to White = 2.30, (M 0.81) = 0.53) SD There was perceived more stigma depression for among Black Combination (M=2.48) Asian (M=2.50), (M=2.50) (M=2.77),Hispanic groups compared to White (M=2.38 and Other (M=2.54) There was stereotype more personal (i.e. attitudes toward depression) compared depression to Blackabout among Asian (M=1.45) Other (M=1.10) Combination (M=0.91), Hispanic (M=1.05), (M=0.93), groups compared to White (M=0.95)

• • • • • • • • Self-report DSS Self-report ISMI PDD Self-report ISMI PDD Self-report PDD Outcomes Community Community Other Students Recruitment =229) =51) =54) =3780 =54) =579) =266) =201) =48) maj maj maj maj min min min min min =302) =240) =290) min min min Hispanic (N Combination (N Other (N Black (N Black (N (N Other Asian (N White (N Black (N White (N Racial groups White (N White (N Type of CMD of Type Various Various Various Depression Country USA USA USA Australia Continued.

Study Conner 2010 Eisenberg Eisenberg 2009 Conner 2009 Copelj 2011 Table 1. 42 ― Chapter 2 Results There was anticipated more stigma depression for related with compared to Whitedepression among Asian SD=1.22) (M=2.45, F=144.40, (p<0.001) SD=1.25) (M=2.10, There was anticipated more stigma depression for related with compared to White SD=1.07) amongemployer Asian (M=2.93, (p<0.001) F=85.55, SD=1.16) (M=2.68, There was anticipated more stigma depression for related with compared to White (M=1.71, family SD=1.19) among Asian (M=2.23, F=360.38SD=1.18) (p<0.001) There was stereotype more personal (i.e. attitudes toward depression) depressionabout among compared to Asian White (M=39.4) P=0.000 (M=15.0) There was anticipated more stigma depression for related with family Hispanic SD=1.24), Black (M=68.5, SD=1.24), among Asian (M=71.7, and Other SD=0.89) (M=60.4, SD=0.96)(M=61.8, groups compared SD=1.00) to White (M=63.1, There was anticipated more stigma depression for related with Black SD=1.08), (M=45.4, family among SD=1.30), Asian (M=55.0, groups SD=1.01) and Other (M=43.01, Hispanic SD=0.91) (M=42.8, SD=1.0) compared to White (M=43.03, There was anticipated more stigma depression for related with African amongemployer 26.8, Asian SD=1.88), (M= (M=42.9, SD=0.92), groups SD=1.01) SD=0.96) and Other (M=27.5, Hispanic (M=28.1, SD=1.0) compared to White (M=27.9,

• • • • • • • • Self-report Study- constructed Vignette Study- constructed Self-report Study- constructed Outcomes Community Community Community Recruitment =100) =1839) =100) =2794) =3596) maj min min min min =841) =3203) =66817) =68319) min maj maj min Asian (N Other (N Asian (N Asian (N Black (N White (N White (N White (N Hispanic (N Racial groups Type of CMD of Type Depression Depression Depression Country USA USA USA Continued.

Study 2005Fogel Georg Hsu 2008 Givens 2007 Table 1. Stigma for Common Mental Disorders ― 43

Results 2 There were no differences in discrimination against those with with those against discrimination in differences no were There depression related with among employer Asian compared to White 1.00) groups (p= with those against discrimination in differences no were There depression related with Asian family (among compared to White 0.05) groups (p= There was discrimination more against those with depression related with friends among Asian compared to White groups (p=0.04) There was discrimination more against those with depression related with doctor/health professional among Asian compared to White groups (p=0.001) There was stereotype more perception those of (i.e. with depression as dangerous) among Asian compared 0.000) to White groups (p= There against blame) was those prejudice more with (i.e. depression among Asian compared 0.000) to White groups (p= There was anticipated more stigma having among any CMDs for groups comparedHispanic Black(12.9%) to Asian (40.3%) (25.9%), p=0.000 compared to White (15.3%)

• • • • • • • Self-report Study- constructed Self-report Study- constructed Outcomes Students Other Recruitment =536) =38) =1257) =184) =112) maj maj min min min =303) min Asian (N (Black (N Asian (N White (N Hispanic (N White (N Racial groups Type of CMD of Type Depression Various Country Australia USA Continued.

Study Hickie 2007 Jimenez 2012 Table 1. 44 ― Chapter 2 Results There were significant no differences in perceivedfor stigma (shame) and Asianhaving between SD=0.67) any CMDs White (M=1.90, t=1.29 SD=0.57), groups (M=2.04, Other perception had group prejudice more migrants of (i.e. with compared to White group depression SE=0.11) as scary) (M=2.28; (p=0.000) SE=0.04)F=8.179; (M=1.82; Other perception had group prejudice more migrants of (i.e. with depression as having problems with comprehension) (M=2.04, F=5.796, SE=0.04) compared to White groups (M=1.64, SE=0.11) (p=0.003) Other feeling had group prejudice more (i.e. uncomfortable) more against migrants SE=0.13) with depression compared to (M=2.50, (p=0.000) F=9.339 SE=0.04) White (M=2.00; Other had group stereotypes more perception migrants of (i.e. with depression SE=0.07) as feeling inadequate (M=2.47, around others) (p=0.029) F=3.539 SE=0.02) compared to White (M=2.31, There was perceived more stigma depression for among Black group t=3.35 SD=18.89) compared to White (M=41.95; SD=12.59) (M=46.16; (p=0.000)

• • • • • • Self-report Study- constructed Self-report Study- constructed LSCS Outcomes Students Community Clinical Self-report Recruitment =1622) =744) =116) =364) =61) =147) maj maj maj min min min Other (N Other Black (N Asian (N White (N White (N Racial groups White (N Type of CMD of Type Various Depression Various Country Germany USA USA Continued.

Study Makowski 2017 Menke 2009 Mokkarala Mokkarala 2016 Table 1. Stigma for Common Mental Disorders ― 45

Results 2 There was internalised more stigma alienation, (i.e. social withdrawal) among the minority Asian compared group to the majority (p=0.615); IN (p=0.161) There was discrimination more desire social more for (i.e. distance) towards those with among CMDs Other compared group to White (p<.001) There was stereotype more those about with among CMDs Other comparedgroup to White (p<.001) There was perceived more stigma depression for among Black compared to White SD=6.44) SD=5.76) groups (M=40.04, (M=42.31, (p=0.000) There was stereotype more personal (i.e. attitudes toward depression) groups and Hispanic depressionabout (p=.30) among Black (p=.037) White. to compared There was stereotype more perceiving (i.e. with people and CMDs groups and Asian (M=11) dangerous) among Black (M=14)(p<.001) (p<.001) and Hispanic (M=9) compared to White (M=12) There was discrimination more desire against segregation) for (i.e. Asian (M=13) (p<.001), those with among CMDs African (M=13) (p<.005) and Hispanic (M=10) groups compared to White (M=11)

• • • • • • • ISMI Self-report CAMI-23 Self-report LSCS Study- constructed Vignette AQ Outcomes Clinical Self-report Community Other Clinical Self-report Students Recruitment =886) =251) =79) =158) =91) =130) =28) =150) =1497) =281) =71) maj maj maj maj min maj min min min min min =5153) =100) min min Asian (N Hispanic (N Asian (N Black (N Black (N (N Other Black (N White (N Hispanic (N White (N Asian (N Racial groups White (N White (N Type of CMD of Type Depression Various Various Various Depression Country USA USA Singapore UK USA Continued.

Study Nadeem 2007 Rao 2007 Picco 2016 Papadopoulos 2002 O`Mahen 2011 Table 1. 46 ― Chapter 2 Results There was stereotype more against those with depression among compared to White (M=105.72, SD=24.74) Asian (M=115.71, (p<0.001) t=4.07 SD=27.08), There was discrimination more desire social for (i.e. distance) among compared to White groups (M=40.26, SD=9.21) Asian (M=37.30, (p=0.001) SD=9.40), t=3.34, There was stereotype more attitudes) against negative those (i.e. compared to Whitewith among groups any CMDs Black (M=2.27) (M=1.93) t=-4.563 (p=0<001) There was prejudice more and discrimination desire for (i.e. againstsegregation) and with among people CMDs Black (p<0.001) Asian groups compared to White (p<0.001) There was less tolerance and with people among support CMDs for and AsianBlack groups (p<0.001) compared to White (p=<0.005) There was discrimination more against those with among CMDs African and Asian groups (p<0.001) compared to White (p<0.001)

• • • • • • Self-report DAQ SDS Self-report CAMI-23 Self-report CAMI-23 Outcomes Students Students Community Recruitment =429) =209) =200) =276) =63) maj maj min min min =2990) maj Black (N Asian (N Comb (N White (N Racial groups White (N White (N Type of CMD of Type Various Depression Various Country UK Canada UK Continued.

Study 2012 Rüsh Shamblaw 2015 Schafer 2011 Table 1. Stigma for Common Mental Disorders ― 47

Results 2 There was discrimination more desire social for (i.e. distance) against those with among SE=0.09) CMDs majority Asian (M=12.00, group compared to minority (M=11.52, SE=0.09), Asian groups (M=10.89, (p<.001) SE=0.45) and Other (M=11.71, SE=0.11) There was perceived more stigma perception those of (i.e. with CMDs as weak among sick) minority not SE=0.06), Asian groups (M=10.95, SE=0.06) (p<.001) compared SE=0.08) to the majority (M=10.07, (M=10.74, There was perceived more stigma perception those of (i.e. with CMDs as dangerous and among unpredictable) minority Asian groups compared to the majority SE=0.11) SE=0.09),(M=11.75, (M=11.60, (p=0.66) SE=0.08) (M=11.61, There was discrimination more desire social for (i.e. distance) against those Other with among any CMDs Black SD=5.04), (M=24.28, compared to SD=4.87) and Hispanic SD=6.23) (M=23.17, (M=23.60, (p=0.000) F=6.32 SD=5.07), White (M=22.41,

• • • • Self-report DSS Vignette SDS Outcomes Community Students Recruitment =467) =32) =65) =977) =963) =1034) =221) maj min min maj min min min =57) min : Sample size Sample : racial for majorities; mean; M: standard SD: deviation; standardp value; SE: p: error; t statistic; t= F: maj Asian (N Hispanic (N Other (N Other Black (N Asian (N Asian (N (N Other White (N Racial groups Type of CMD of Type Various Various Country Singapore USA Continued.

Various: Several are CMDs Various: studied the type together and/or was specified; not CMD of Common Mental CMDs: Disorders; Recruitment: Community: : Sample size Sample : racial for minorities; N min Study Subramaniam 2017 Wang 2013 2013 Wang Table 1. Note. Community Clinical: sample; Clinical Attribution Student: sample; Student Questionnaire; AQ: sample; Community CAMI-23: Attitudes towards Mentally Ill Scale; Internalized ISMI: Stigma Mental of Illness Perceived Scale; Devaluation PDD: and Discrimination Scale; Social SDS: Distance Scale; DSS: Depression DepressionStigma Scale; Attribution DAQ: Questionnaire: LSCS: Link Stigma Consciousness Scale; Study-Constructed: study-constructed questionnaires; N F statistic. F 48 ― Chapter 2

Risk of bias The risk of bias can be seen in table 2. When taking into account the five different bias items, 20 studies (68%) were rated as high risk of bias, 6 studies (21%) were rated as moderately high risk of bias and only 3 (10%) studies were rated as low risk of bias. The selection bias was rated as low risk of bias in 10 studies (34%), high risk of bias in 18 studies (62%), and unclear in 1 study (3%). When taking into account the study design, there was a high risk of bias in in 18 studies (62%) and low risk of bias in 11 studies (38%). Data collection methods were rated as low risk of bias in 10 studies (34%), high risk of bias in 16 studies (55%) and unclear in 3 studies (10%). The risk for confounders was high in 20 studies (68%) and low in 9 studies (31%). As for the data analyses, the risk of bias was low in 22 studies (76%), high in 1 study (3%) and unclear in 6 studies (21%). We assessed whether the authors used stratification and/or matching in the study design in order to control possible confounders (Lavrakas, 2008; Keeble et al., 2015). Only 4 studies (14%) used stratification. Further, the representativeness of the samples was also limited as only 4 studies (14%) used random sampling, whereas the rest of the studies used convenience sampling. In one study, the authors recruited participants from their friends and networks of their friends which increased the chances of selection bias and restricted the representativeness of the racial groups in their sample (Georg Hsu et al., 2008). Out of the 21 self-report studies, 7 (33%), and out of the 8 vignette studies, 3 (37%) used “study constructed’’ questionnaires (i.e. questionnaires which are constructed by the author for the purpose of the study) as outcome measures. The psychometric properties of these questionnaires were not tested. Stigma for Common Mental Disorders ― 49

Table 2. Risk of Bias Assessment of all the studies (N = 29)

Type of Data Select Study Study Studies Analyses Confound collection Bias Design RoB a) Adewuya, 2008 Self-report Low High Low Low High High Ahn 2015 Self-report Low High Low Low High High Anglin 2006 Vignette Low Low High Low Low High Aznar-Lou 2016 Self-report Low Low Low High Low Moderate Brown 2010 Self-report High Low High Low Low High Caplan 2011 Self-report Low High High High High High 2 Cheng 2015 Vignette Low Low Unclear Low Low Moderate Conner 2010 Self-report Low High Low Low Low Moderate Conner 2009 Self-report Low High Low High Low High Copelj 2011 Self-report Low High High High Low High Eisenberg 2009 Self-report Low Low Low High Low Moderate Fogel 2005 Self-report Low High High Low Low High Georg Hsu 2008 Vignette Unclear High High High High High Givens 2007 Self-report High High Low High Low High Hickie 2007 Self-report Low High High High Low High Jimenez 2012 Self-report Low Low High Unclear Low Moderate Makowski 2017 Self-report Low Low High Low Low Moderate Menke 2009 Self-report Low High Low High High High Mokkarala 2016 Self-report Low High High Low Low High Nadeem 2007 Self-report Low Low High High Low High O`Mahen 2011 Self-report Low Low Unclear High High High Papadopoulos 2002 Self-report Low High Unclear High High High Picco 2016 Self-report Low Low Low High High High Rao 2007 Vignette Unclear Low Low High Low High Rüsh 2012 Self-report Low Low Low Low Low Low Schafer 2011 Self-report High High Low High High Low Shamblaw 2015 Self-report Low High Low High High High Subramaniam 2017 Vignette Low Low Low Low Low Low Wang 2013 Vignette Low High High High High High Note. Low: Low risk of bias; High: High risk of bias; Unclear: reviewers were not able to reach consensus due to lack of information; a) In this column, high refers to the high risk of bias (studies which scored high risk of bias in 3 or more of the assessment domains); moderate refers to the moderate risk of bias (studies which scored high risk of bias in 2 of the assessment domains) and low refers to the low risk of bias (studies which scored high risk of bias in 1 of the assessment domains); RoB: Risk of Bias Assessment 50 ― Chapter 2

Stigma for CMDs Between Racial Minorities and Majorities First, we run the analyses separately for the studies investigating depression only g=0.22 (95% CI: 0.10~0.34) (I2=94%, 95% CI: 93~96) and for the studies investigating various types of CMDs together g=0.18 (95% CI: 0.10~0.28) (I2=86%, 95% CI: 81~90). Since this resulted with small effect sizes and with very high heterogeneity in each, we decided to pool all the studies together, regardless of the condition of the mental illness studied, for further sensitivity analyses. Primary outcome was mental illness stigma defined by the MISF (Fox et al., 2018). Cognitive (e.g. stereotype), affective (e.g. prejudice), behavioural (e.g. discrimination) and/ or combination of each of these components of mental illness stigma were compared between racial minorities and majorities in 29 studies (39 comparisons). We have decided to pool the studies together as stigmatiser perspective measuring the cognitive component of the stigma was over represented, whereas stigmatised perspective measuring how stigma was anticipated was under represented (see table 1). The overall effect size was small but significant g=0.20 (95% CI: 0.12~0.27) indicating that racial minorities had more mental illness stigma about CMDs when compared with the majorities, with very high heterogeneity (I2=91%, 95% CI: 89~93). The results of the analyses are reported in table 3, and the forest plot is given in table 5. When studies with high risk of bias (i.e. defined as those with high risk of bias scores on three and/or more categories of the risk of bias assessment tool) were excluded, the effect was sustained g=0.20 (95% CI: 0.10~0.25) and the heterogeneity was still very high (I2=88%, 95% CI: 84~90). When only studies with standardized outcome measures were included, the effect size was slightly greater g=0.23 (95% CI: 0.10~0.36) and the heterogeneity remained high (I2=84%, 95% CI: 75~89). Next, excluding outliers resulted in a comparable effect size g=0.29 (95% CI: 0.21~0.36) and less heterogeneity (I2=29%, 95% CI: 0~63). To identify outliers, we looked at whether the 95% CI of the study overlaps with the 95% CI of the pooled effects size and we also looked whether a study considerably differs from the other included studies in the metanalysis (Cuijpers et al., 2016). Even though the CI of the study by Subramaniam and colleagues (Subramaniam Stigma for Common Mental Disorders ― 51 et al., 2017) did fall between the CI of the pooled effect sizes, the study did conceptualise stigma differently compared to the other studies. Additionally, excluding that study reduced the heterogeneity significantly and therefore we decided to exclude it (see table 3). Egger’s test did not indicate a significant publication bias (p=0.100). Duvall and Tweedie’s trim and fill procedure indicated 2 missing studies. The adjusted effect size (after imputation of the missing studies) was g=0.27 (95% CI: 0.20~0.33). We did not investigate whether there was a variability in stigma perspectives among racial minorities in the subgroup and in multi-variate analyses since the remaining 2 number of studies in some categories (perceived stigma N=2, stigmatised N=1) were not sufficient. No significant differences were found in subgroup analyses (see table 2). Multi-variate analyses indicated no significant associations between the effect size and the racial groups (p=0.42), quality (p=0.12) and the type of the outcome of the studies (self-report, vignette) (p=0.48) (see table 3). However, there was a significant association between studies with high risk of bias and the effect size of stigma (p=0.04). 52 ― Chapter 2

Table 3. Stigma for racial minorities and majorities: Pooled effect sizes of primary outcomes

2 a Characteristics Ncomp g 95% CI I 95% CI P Primary Analyses

All analyses 39 0.20 0.12~0.27 91% 89~93 <.001 High risk of bias studies excluded 29 0.20 0.10~0.25 88% 84~90 <.001 Standardized outcomes only 16 0.23 0.10~0.36 84% 75~89 <.001

Outliers excluded 12 0.29 0.21~0.36 29% 0~63 <.001

Subgroup Analyses

Ethnicity Asian 4 0.29 0.17~0.42 23% 0~75 <.001 Black 5 0.30 0.13~0.36 55% 0~81 <.001 Other 3 0.26 -0.01~0.41 17% 0~77 . 001 Total betweenb .93

Quality Moderate 5 0.21 0.10~0.32 0% 0~64 <.001 Strong 2 0.24 0.13~0.34 a) b) <.001 Weak 5 0.40 0.26~0.54 33% 0~75 <.001 Total between .08

Outcome Self-report 7 0.25 0.14~0.36 37% 0~72 <.001 Vignette 5 0.34 0.25~0.44 0% 0~64 <.001 Total between .19

a Note. Ncomp: Number of comparisons; P : Values indicating the difference within subgroups; Total betweenb: p value indicating the difference between the sub groups a) The 95% PI cannot be calculated when the number of studies is lower than 3. b) The 95% CI of I2 cannot be calculated when the number of studies is lower than 3. Stigma for Common Mental Disorders ― 53

Table 4. Multi-variate meta-regression analyses of predictors of stigma, by quality of studies, ethnicity and type of stigma outcomes in 10 studies of stigma in ethnic minorities and majorities

Characteristics N SE β 95% CI ZP Stigma Outcome Self-report 7 0.09 0.14 -0.17~0.37 0.70 .48 Vignette (ref) 5

Quality of the Studies .11 2 Weak 5 0.13 0.27 0.01~0.53 2.07 .04 Strong 2 0.11 -0.03 -0.25~0.18 -0.34 .73 Moderate (ref) 5

Ethnicity .42 Asian 4 0.12 0.28 -0.08~0.41 1.31 .19 Black 5 0.10 0.53 -0.11~0.29 0.83 .40 Other (ref) 3

Note. Point Est: Point Estimate; p: values indicating the difference between the effect sizes in subgroups; ref: reference group 54 ― Chapter 2

Table 5. Stigma for CMDs between racial minorities and majorities: Effect sizes of primary outcomes in all studies

Ethnic Forest plot of Hedges’ Study Group Type Outcomes g 95% CI g and 95% CI Adewuya, 2008 [38] Other SR SDS -0.15 -0.36~0.05 −0.15 ( Ahn 2015 [39] Asian SR PDD -0.12 -0.21~-0.03 −0.12 ( Anglin 2016 [18] Black VIG Study -0.13 -0.32~0.05 0.66 ( 0.53 to −0.130.81 ) ( Aznar-Lou 2016 33] Asian SR CAMI-23 0.67 -0.06~1.42 0.67 ( − Aznar-Lou 2016 33] Black SR CAMI-23 0.15 -0.11~0.41 0.15 ( − Aznar-Lou 2016 33] Other SR CAMI-23 0.17 -0.07~0.41 0.17 ( − 1.36 ( 0.87 to 2.14 ) Brown 2010 [19] Black SR ISMI, PDD -0.07 -0.25~0.11 −0.07 ( Caplan 2011[22] Hisp. SR Study -0.04 -0.33~0.24 −0.04 ( Cheng 2015 [15] Asian VIG AQ 0.29 0.10~0.48 0.29 ( 0 Conner 2010 [43] Black SR ISMI, PDD 0.16 -0.02~0.35 1.36 ( 1.05 to 0.161.77 () − Conner 2009 [40] Black SR ISMI, PDD 0.76 0.36~1.17 0.76 ( 0 Copelj 2011[41] Other VIG DSS 1.04 0.64~1.44 1.04 ( 0 Eisenberg 2009 [42] Asian SR PDD 0.16 0.07~0.24 0.16 ( 0 Eisenberg 2009 [42] Black SR PDD 0.51 0.39~0.64 2.04 ( 1.38 to 0.512.99 () 0 Eisenberg 2009 [42] Hisp. SR PDD 0.15 0.04~0.27 0.15 ( 0 Eisenberg 2009 [42] Other SR PDD 0.17 0.04~0.29 0.17 ( 0 Fogel 2005 [16] Asian SR Study 0.31 0.26~0.35 0.31 ( 0 Georg Hsu 2008 17] Asian VIG Study 0.67 0.27~1.07 2.32 ( 1.69 to 0.673.2 ) ( 0 Givens 2007 [21] Native SR Study -0.21 -0.27~-0.14 −0.21 ( Givens 2007 [21] Hisp. SR Study -0.01 -0.05~0.02 −0.01 ( Hickie 2007 [43] Asian SR Study 0.33 -0.01~0.68 0.33 ( − Jimenez 2012 [23] Black SR Study 0.16 0.05~0.26 2.36 ( 2.04 to 0.162.73 () 0 Makowski 2017 11] Other VIG Study 0.15 -0.00~0.31 0.15 ( 0 Menke 2009 [44] Black SR LSCS 0.23 0.05~0.41 0.23 ( 0 Mokkarala 2016 45] Asian SR Study 0.21 -0.09~0.52 0.21 ( − Nadeem 2007 [8] Black SR Study 0.16 -0.09~0.52 2.43 ( 1.87 to 0.163.16 () − Nadeem 2007 [8] Hisp. SR Study 0.13 0.01~0.24 0.13 ( 0 O`Mahen 2011 [46] Black SR LSCS 0.37 0.20~0.54 0.37 ( 0 Papadopoulos 2002 47 Other SR CAMI-23 0.51 0.09~0.24 0.51 ( 0 Picco 2016 [48] Asian SR ISMI -0.14 0.20~0.81 2.43 ( 1.83 to −0.143.23 ) ( Rao 2007 [49] Asian VIG AQ 0.47 0.19~0.75 0.47 ( 0 Rao 2007 [49] Hispanic VIG AQ 0.42 0.17~0.67 0.42 ( 0 Rüsh 2012 [57] Asian SR CAMI-23 0.21 0.11~0.45 0.21 ( 0 2.89 ( 2.1 to 3.98 ) Rüsh 2012 [50] Black SR CAMI-23 0.28 0.22~0.89 0.28 ( 0 Schafer 2011[51] Black SR CAMI-23 0.56 0.16~0.54 0.56 ( 0 Shamblaw 2015 [52] Asian SR DAQ, SDS 0.35 -0.23~-0.06 0.35 ( − Subramaniam 2017 [23] Asian VIG DSS -0.14 0.08~0.23 3.01 ( 1.72 to −0.145.26 ) ( Wang 2013 [53] Black VIG SDS 0.36 0.20~0.53 0.36 ( 0 Wang 2013 [53] Hisp. VIG SDS 0.15 -0.12~0.42 0.15 ( − 12340.00.0 0.5 0.5 1.0 1.0 5 1.5 1.5 Stigma for Common Mental Disorders ― 55 7 5 7 9 6 6 7 2 8 6 5 − − − 0 − 0 0 0 0 0 0 0 0 − 0 0 0 − − 0 0 0 0 0 0 0 0 − 0 − ( ( ( ( ( ( ( ( ( 0.6 0.1 0.1 0.2 0.1 0.7 0.4 0.4 0.2 0.2 0.3 0.1 −0.15 −0.12 −0.13 0.67 ( 0.15 ( 0.17 ( −0.07 −0.04 0.29 ( 0.16 ( 0.76 ( 1.04 ( 0.16 ( 0.51 ( 0.15 ( 0.17 ( 0.31 ( 0.67 ( −0.21 −0.01 0.33 ( 0.16 ( 0.15 ( 0.23 ( 0.21 ( 0.16 ( 0.13 ( 0.37 ( 0.51 ( −0.14 0.47 ( 0.42 ( 0.21 ( 0.28 ( 0.56 ( 0.35 ( −0.14 0.36 ( 0.15 ( 0.66 ( 0.53 to 0.81 ) 1.36 ( 0.87 to 2.14 ) 1.36 ( 1.05 to 1.77 ) 2.04 ( 1.38 to 2.99 ) 2.32 ( 1.69 to 3.2 ) 2.36 ( 2.04 to 2.73 ) 2.43 ( 1.87 to 3.16 ) 2.43 ( 1.83 to 3.23 ) 2.89 ( 2.1 to 3.98 ) 3.01 ( 1.72 to 5.26 ) 5 0.0 0.5 1.0 1.5 g (95% CI) g (95% 1234 0.0 0.5 1.0 0.0 0.5 0.0 0.5 1.0 CI 2 -0.12~0.42 0.22~0.89 0.11~0.45 0.19~0.75 0.36~1.17 -0.02~0.35 0.10~0.48 -0.07~0.41 -0.06~1.42 -0.11~0.41 95% 0.15 0.36 0.20~0.53 0.28 0.21 0.42 0.17~0.67 0.47 0.76 0.16 0.29 0.17 0.67 0.15 g SDS SDS CAMI-23 CAMI-23 AQ AQ ISMI, PDD ISMI, PDD AQ CAMI-23 CAMI-23 CAMI-23 Outcomes vignette vignette Self-report Self-report vignette vignette Self-report Self-report Vignette Self-report Self-report Self-report Type of Study of Type Hispanic Black Black Asian Hispanic Asian Black Black Asian Other Asian Black Ethnic Group Stigma between CMDs for racial minorities and outliers when majorities: plot Forest are excluded Wang 2013 [53] 2013 Wang Wang 2013 [53] 2013 Wang Rüsh 2012 [50] 2012 Rüsh Rüsh 2012 [50] 2012 Rüsh Rao 2007 [49] Rao 2007 [49] Conner 2009 [43] Conner [43] 2010 Cheng 2015 [15] Aznar-Lou 2016 [33] Aznar-Lou 2016 [33] Study Aznar-Lou 2016 [33] Table 6. 56 ― Chapter 2

discUssion This systematic review and meta-analysis was aimed at comparing mental illness stigma associated with CMDs based on racial background. In line with our expectations, the results suggest that racial minorities have more mental illness stigma for CMDs when compared with racial majorities. Sensitivity analyses showed the robustness of these results. The multi-variate meta-regression analyses indicated that studies of poor quality had higher effect sizes than the studies with high quality. Higher mental illness stigma in racial minorities is in line with the growing literature highlighting the variations in mental illness stigma based on ethnicity and/or race (Makowski et al., 2017; Corrigan et al., 2007; Fogel et al., 2005; Conner et al., 2009; Subramaniam et al., 2017; Miranda et al., 2015). This finding could be explained in relation to the social identity theory (Hogg & Smit, 2007). In collectivistic cultures, group harmony and cohesion are of central importance (Bhui, 2019). In this context, CMDs might be seen to fall outside of societal expectations (Bhui, 2019; Hogg & Smit, 2007; Crenshaw, 1989) and this would precipitate shame and subsequent stigma (Hogg & Smit, 2007; Augsberger et al., 2015). Thus, what is defined as “them” not “us”, would reinforce the public opinion that individuals with CMDs are dangerous and must be segregated (Jorm & Oh, 2009). Consistent with this notion, there is research evidence indicating that cultural beliefs of CMDs are related with the extent of the impact of mental illness stigma (i.e. fear of someone with CMDs and/or desire for social distance from an individual with CMDs) at individual and societal levels (Bhui, 2019; Jorm & Oh, 2009; Shefer et al., 2012; Crenshaw, 1989). In extent, the consequences of mental illness stigma are more harmful especially when the preceding cultural and/or personal attributions coincide with social adversities such as migration, poverty, gender, diagnosis of a CMD, ethnic and/or sexual minority statuses (Korszun et al., 2012; Lee et al., 2009; Shefer et al., 2012; Brenner, 2019). This phenomenon is known as the intersectional impact of stigma and it accounts for the underutilisation of mental health services among those who are affected (Clement et al., 2015; Brenner, 2019). For instance, a large scale study from America (N=15,383) showed Stigma for Common Mental Disorders ― 57 that, Black women with low socio-economic background reported more stigma related concerns and were less likely to utilise mental health services compared to locally born women from White backgrounds (Nadeem et al., 2007). Furthermore, the link between intersectionality and underutilisation of health services is well-documented in the HIV literature investigating the impact of the intersection of different types of stigma (HIV-related stigma, sexism, racism, and homo/transphobia) on individuals` well-being (Henkel et al., 2008; Radcliffe et al., 2010; Logie et al., 2011). Studies comparing public beliefs and attitudes regarding CMDs point to the 2 importance of migration status and/or ethnicity in shaping mental illness stigma (Bhui, 2019). For instance, in a large scale vignette study (N=2013) by Makowski and colleagues in Germany, there was an indication for the differences in mental illness stigma between locally born minorities, migrants and non-minorities (Makowski et al., 2017). When the vignette concerned an individual from a migrant background with depression, participants from migrant background expressed greater stigmatizing attitudes (e.g. negative stereotypes, emotional reactions and desire for social distance) compared to the locally born minorities and majorities (Makowski et al., 2017).

Strengths and limitations This study has various innovations. This is the first meta-analysis in the stigma field utilising a unified conceptual framework to pool the studies. This is an important step to advance the current understanding of the mental illness stigma and how racial/ethnic differences impact people`s experiences of mental illness stigma. Based on the current results an important message for the public health field is to tailor the existing anti-stigma interventions according to the specific racial and/ or ethnic groups (Gronholm et al., 2019). At present, recommended practices include psycho-educational campaigns aiming to improve the public knowledge on CMDs and individuals with CMDs (Gronholm et al., 2019; Corrigan, 2016; Dumensnil & Verger, 2009). Such practices however, fail to demonstrate changes in attitudes among stigmatised and stigmatiser (Dumensnil & Verger, 2009). Often, educational campaigns 58 ― Chapter 2 do not provide evidence for the effectiveness of anti-stigma campaigns in increasing help- seeking for CMDs (Dumensnil & Verger, 2009). Our results suggest that the development and the implementation of such campaigns can be improved if the messages of these campaigns are adapted according to the socio-cultural and political contexts of the countries that the individuals live in (Dumensnil & Verger, 2009; Corrigan, 2016; Bhui, 2010a; 2010b). Even though our results suggest that all racial minorities, regardless, have higher mental illness stigma, various considerations need to be taken into account whilst interpreting our findings. We found no variability in stigma differences based on specific races. This could be explained with the small number of studies representing each racial minority group in sub-group analyses. Further, we were also unable to investigate if there were differences in mental illness stigma between racial minorities based on the migration status since this was rarely investigated in the included studies (Nadeem et al., 2007; Lee et al., 2009; Clement et al., 2015; Makowski et al., 2017). It may be that there are other potential moderators such as the degree of acculturation which are overlooked in the current stigma literature and could not be taken into consideration in our study. Further, the multi-variate analyses pointed to the possible moderating role of the number of studies with high risk of bias on the effect size of stigma. It could be that if there were more studies with low risk of bias, we might have found no racial differences in stigma for CMDs. Given these reasons, the results of this study must be treated with caution. The studies from North America were overrepresented in the current study. This limits our understanding of how contextual differences between continents influence the impact of mental illness stigma. For instance, the experiences of discrimination related with stigma might be different in Europe compared to America (Bhui, 2019). Related issues were the racial classification and the definition of minority and/or majority statuses in this study. Since there were many ethnic and religious groups in the included studies, we decided that the mutually exclusive category approach (Conelly et al., 2016), re-allocating individuals in existing categories defined by the previous research Stigma for Common Mental Disorders ― 59

(Ünlü İnce et al., 2014), would be the most suitable. We have also defined minority and majority statuses based on the country of the included studies. Since the definition of race-ethnicity includes history, religion, language and socio-political dynamics (Conelly et al., 2016; Bhui, 2010a), we oversimplified the term by using broad racial categories. Moreover, our definitions do not capture the changes in ethnicity and minority statuses over time depending on the changing socio-political circumstances (Conelly et al., 2016). An alternative to the preceding is the multiple characteristics approach (Conelly et al., 2016) taking into account the various aspects of ethnicity such as language, country of 2 birth, nationality and religiosity (Conelly et al., 2016). Even though the latter offers more effective approach to the measurement of ethnicity in detail, the former is more pragmatic and facilitates the comparability of the studies across countries (Conelly et al., 2016). In light of these caveats, our results are restricted with the broad racial categories which are defined at a specific point in time. It is recommended that ethnicity and minority statuses are measured at different time points in prospective studies (Conelly et al., 2016). Another limitation restricting the representativeness of the sample in our study was the exclusion of clinical populations that had comorbidities. One reason could be that we have excluded HIV, and other comorbidities and yet most clinical studies include people who often have CMDs and other conditions. Some studies indicate that diagnosis of a CMD would reinforce the label indicating that the individual falls out of the societal expectations and therefore is unpredictable and/or dangerous (Nadeem et al., 2007; Shamblaw, Botha & Dozois, 2015). Alternatively, diagnosis of a CMD might create opportunities for more psycho-education about the CMDs, and more contact with the others with CMDs and this might reduce the mental illness stigma in return (Corrigan, 2016). Since clinical populations are not represented in our study, we were not able to assess if there was a variation in mental illness stigma among racial minorities depending on the presence of a formal diagnosis. Thus, the following questions are important and yet they remain to be answered by the future studies: Among those who have received a diagnosis of a CMD, are there any differences in the experience of mental illness stigma between ethnic minorities and majorities? Among ethnic minorities, are there differences 60 ― Chapter 2 in mental illness stigma between those who have been diagnosed with CMDs and those who have not? Further, some types of stigma are under investigated. There is an indication that stigma is associated with greater burden when it is accepted and internalized by those who have a mental illness (Makowski et al., 2017). In our meta-analysis, many studies investigated stereotypes or prejudices attributed to those who have a mental illness. Conversely, how stigma was anticipated or experienced by racial minorities when compared with majorities were rarely investigated (Caplan et al., 2010; Jimene et al., 2013). Given this limitation, we could not investigate whether there was a variation in different perspectives (e.g. anticipated, internalised) and/or components (e.g. cognitive, affective, behavioural) of mental illness stigma between racial minorities. It is worth to emphasise that there were considerable number of the questionnaires which were developed for the purpose of the studies by the authors themselves. In our sensitivity analyses, exclusion of these questionnaires and limiting the analyses with the validated questionnaires such as CAMI, reduced the heterogeneity. The revision of the existing self-report questionnaires by the authors was another limitation. Often, these questionnaires were revised according to the research question and/or the sample and their psychometric properties were not tested. Additionally, the self-report questionnaires often measured multiple stigma mechanisms within the same scale (Fox et al., 2018). This might conflate the results as stigma mechanisms might differently relate to the outcomes. Poor quality of the vignette studies also worth mentioning. The vignettes were developed based on the DSM criteria and the authors` clinical information of the CMDs. Often, there was no information about the psychometric properties of the vignettes with one exception. Subramaniam and colleagues followed a systematic approach by revising the vignettes with clinical experts, piloting them with participants and revising the relevance and acceptability afterwards (Subramaniam et al., 2017). Since the vignettes are often developed by the researchers themselves, it could be that they are more suggestive when compared with self-report measures and this would confound the results (see appendix B). Stigma for Common Mental Disorders ― 61

Our results are also limited with the MISF as other frameworks exist such as the Framework Integrating Normative Influences on Stigma (FINIS) (Pescosolido et al., 2008). We have chosen the MISF as to our knowledge, it identifies specific outcome measures which are capturing specific stigma mechanism (Fox et al., 2018).

Implications and conclusions The limitations of our study underscore the importance of investigating the intersection of race/ethnicity, degree of acculturation, presence of a mental illness diagnosis and 2 the impact of mental health stigma. There is a need for more high quality research for the advancement of the stigma field. The quality of the future studies could be improved by defining meaningful controls. For instance, researchers could relate sample characteristics to the general population of the country of the study. Additionally, the racial and or ethnic groups of the country of the studies should be represented. In line with the multiple characteristics approach to define ethnicity, future studies could define the ethnic composition of their sample consistently based on the characteristics such as history and religion which are outlined earlier. Furthermore, more prospective studies are needed to capture the changes in ethnic classification over time. Prospective studies are also crucial to examine whether the degree of acculturation, diagnosis of a CMD and/or minority statuses are effect modifiers in the relationship between ethnicity and mental illness stigma. To conclude, mental illness stigma is one of the important myriad of factors that might underpin individuals` state of physical, psychological, and social wellbeing. The results of the current meta-analysis indicate differences in mental illness stigma based on racial background and this result highlights the important role of racial and/ethnic background in shaping the mental illness stigma. An important clinical implication of these findings would be to tailor anti-stigma strategies according to the specific racial and/or ethnic backgrounds with the intention to improve mental health outreach (Gronholm et al., 2019).

3

accUltUration and sUicidal ideation among tUrkish migrants in the netherlands

pUblished as Eylem, O., Dalgar, I., Tok, F., Unlu Ince, B., van Straten, A., de Wit, L., Kerkhof, A.J.F.M., Bhui, K. (2019). Acculturation and suicidal ideation among Turkish migrants in the Netherlands. Psychiatry Research, 275,71-77, https:// doi.org/10.1016/j.psychres.2019.02.078 64 ― Chapter 3

abstract

Background More suicidal ideation and higher rates of attempted suicide are found in Turkish people when compared with the general population in Europe. Acculturation processes and related distress may explain an elevated risk of suicide. The current study investigates the association between acculturation and suicidal ideation among Turkish migrants in the Netherlands. The mediating effect of hopelessness and moderating effect of secure attachment are also examined.

Methods A total of 185 Turkish migrants living in the Netherlands were recruited through social media and through liaison with community groups. They completed an online survey including validated measures of suicidal ideation, hopelessness, acculturation and attachment style. Mediation and moderation analyses were tested using bootstrapping.

Results Higher participation was associated with less hopelessness and less suicidal ideation (B = -0.07, Boot SE = 0.03, p< 0.001, 95% CI [-0.14, -0.02]). Greater maintenance of one’s ethnic culture was associated with higher hopelessness and higher suicidal ideation (B = 0.03, Boot SE = 0.02, p< 0.001, 95% CI [0.01, 0.07]). Greater participation was associated with less suicidal ideation particularly amongst those with less secure attachment styles (B = -0.32, SE = 0.05, p < 0.001, 95% CI [-0.42, -0.22]).

Conclusions Turkish migrants who participate in the host culture may have a lower risk of developing suicidal thinking. Participation may protect against suicidal thinking, particularly among those with less secure attachment styles. Acculturation and Suicidal Ideation ― 65

3 66 ― Chapter 3

introdUction

here is a variation in rates and characteristics of suicidal behaviours between and within migrants in Europe. Some of this variation is associated with certain ethnic groups being at special risks for suicidal behaviours (Haigh et al., 2016). Turkish populations are one Tof the largest migrant populations in Europe whose suicidal behaviour deserves further attention (Bursztein Lipsicas et al., 2012). Gender and age disparities exist between the Turkish migrants and the native populations from across Europe in the manifestation of suicidal behaviours (Aichberger et al., 2015b). In the Netherlands, people of Turkish descent die because of suicide at a younger age compared to Dutch natives (Turkish men: mean age 32; Turkish women: 32; Dutch men: 48; Dutch women: 51) (Garssen et al., 2006). Turkish immigrant girls (age 10-17) in Germany die more often by suicide than Turkish immigrant males, and the female to male ratio (3:2 among German vs 0:6 among Turkish) is reversed compared to the native German population in this age group in particular (Razum & Zeeb, 2004). There is also an increased risk of suicidal ideation in Turkish adolescents (38.1%) compared to Moroccan (28.9%) and Dutch (17.9%) adolescents (van Bergen et al., 2008). Furthermore, the hospital based registration studies showed that young women of Turkish descent in Germany (Lizardi et al., 2006), in the Netherlands (Burger et al., 2006) and in Switzerland (Brückner et al., 2011) respectively, are more than five times more likely to attempt suicide than German native women (0.51% vs 0.09%) two-three times more likely to attempt suicide than Dutch native women (0.54% vs 0.25%) and three-four times more likely than Swiss native women (0.65% vs 0.17%). Currently, little is known about how the risk for suicidal behaviours is generated among Turkish migrants. Since migration background is also relevant to other migrant groups, it is perhaps not crucial in explaining the increased suicide risk in Turkish migrants in particular. Acculturation and Suicidal Ideation ― 67

Difficulties in acculturation processes might be central to the increased suicidal ideation and attempts (Bhugra, 2004; van Bergen et al., 2012). Acculturation occurs when a culture encounters a dominant alternative culture (Berry, 1997; Brown & Zagefka, 2011). Berry`s bi-dimensional model of acculturation suggests that at the psychological level, ethnic groups and individuals involved in the acculturation process face two fundamental issues: The first issue, “cultural maintenance” or bonding social capital, is the wish to linkage to those of one’s cultural background and the second issue, “participation” or bridging social capital, is the wish to linkage to those of other majority and minority groups (Brown & Zagefka, 2011). 3 How the acculturation experience of an ethnic minority group becomes stressful can be explained through the culture conflict theory (Bhugra, 2004; Lester, 2008; Lester, 2013). When the perceived pressure to adjust to the majority culture contrasts with the expectations for cultural continuity in minority culture, these stressful experiences can precipitate suicidal behaviours among members of the minority group. Research in Belgium indicates that Turkish migrants with greater cultural maintenance report more distress in their contact with other minority and majority populations especially when they perceive their ethnic identity as incompatible with the majority culture (Güngör, 2008; Güngör & Bornstein, 2009; Phalet and Schonpflug, 2001). It is known that Turkish migrants in Europe have strong ties with their country of origin (Klok et al., 2017). In line with the cultural conflict theory, it could be argued that greater cultural maintenance among Turkish migrants might leave them more prone to experience distress in their daily interactions with other groups which may precipitate other suicide risk factors such as hopelessness leading to suicidal behaviours (Baysu et al., 2013; Phalet et al., 2015; van Acker & Vanbeselaere, 2012). Hopelessness is one of the most heavily studied suicide risk factors (e.g. Lester et al., 1979). It is a cognitive component of depression and is defined as a decrease in positive future expectancies (van Beek, 2013). Recent research suggests ethnic variability in relation between hopelessness and suicidal behaviours (Polanco-Roman & Miranda, 2013; Durant et al., 2006). For 68 ― Chapter 3 instance, a prospective study on an ethnically diverse sample of migrants indicated that hopelessness was a mediator in the associations between culturally related stressors and suicidal ideation (Polanco-Roman & Miranda, 2013). In extend, the association between acculturation and suicidal ideation might be moderated by other suicide risk factors such as attachment (e.g. Zeyrek et al., 2009, van Leeuwen et al., 2010). Attachment refers to the interaction strategies with others in new situations and is derived from the early emotional experience with primary caretakers (Bowlby, 1973; Bartholomew et al., 1991; Lizard et al., 2011). There is well-established evidence suggesting a link between insecure attachment styles and suicidal behaviours (e.g. Adam et al., 1996; de Jong, 1992; Zeyrek et al., 2009; Stepp et al., 2008). Attachment has been identified as one of the predictors of particular acculturation dimensions (Bhugra, 2004; van Leeuwen et al., 2010). Theoretically, it could be argued that, attachment styles in early years of life form the basis of our attitudes and adaptation strategies in our in-group and out-group interactions (Hofstra et al., 2005; van Oudenhoven & Hofstra, 2006). Further support for the predictive role of attachment in acculturation comes from both correlational and prospective studies. For instance, migrants with secure attachment styles were found to be positively oriented to participating in the host country whereas; migrants with dismissive-avoidant attachment styles, appeared to be more positively oriented to maintaining their heritage culture (Handojo, 2000; De Pater et al., 2003; Bakker et al., 2004). In the present study, we assess the extent of the relationship between acculturation and suicidal ideation among the Turkish migrants in the Netherlands. On the basis of the attachment theory, we assume that attachment style is settled early in life and is an effect modifier increasing or decreasing the risk for suicidal behaviours. Hopelessness on the contrary, is perhaps dependent on the other culture related stressors such as, acculturation (e.g. Polanco-Roman & Miranda, 2013) and functions as a mediator in the associations between the acculturation and suicidal ideation. It is expected that, a more secure attachment style, and greater participation in the host culture is associated with Acculturation and Suicidal Ideation ― 69 less hopelessness and less suicidal ideation. On the contrary, a less secure attachment style, and greater maintenance of one’s ethnic culture is associated with more hopelessness and more suicidal ideation. We investigate the following research questions:

• Is there an association between acculturation and suicidal ideation? • Does hopelessness mediate the relationship between acculturation and suicidal ideation? • Does attachment style is an effect modifier of the association between acculturation and suicidal ideation? 3 method

Participants Participants were 185 (114 women, 71 men) Turkish migrants with a mean age of 36 years (M=36, SD = 11.9, range 18-75). The aim was recruiting 1st and 2nd generation individuals with Turkish descent. The inclusion criteria were: living in the Netherlands, having Turkish ethnic background (i.e. with at least one parent or grandparent born in Turkey), 18 and older, having sufficient command of Dutch or Turkish language and giving informed consent.

Procedure The study was approved by the Scientific and Ethical Review Board of the Faculty of Psychology and Education of the VU University Amsterdam (VCWE).The recruitment took place in 2013 and 2014 based on convenience sampling. Participants were recruited through social media and through liaison with community groups. Interested participants contacted the research team through e-mail. Then the information letter, informed consent form and a link to the online questionnaire were shared. Both Turkish and Dutch languages were used during the inclusion process. 70 ― Chapter 3

Materials Participants could choose between the Turkish and the Dutch versions of the questionnaires. Only 8 participants chose the Dutch language. Thus, we couldn`t compare those who filled in the Dutch questionnaire with those who filled in the Turkish questionnaire to see if there was any effect of the language choice (Oyserman & Lee, 2008). Suicidal ideation was measured through the suicide ideation subscale of the Suicide Probability Scale (SPS; Cull & Gill, 1990) which is a self-report measure for the assessment of an attempted suicide potential in adolescents and adults. The suicide ideation subscale consists of 8 items. Items are scored using a four points scale where 1 refers to ‘none or a little of the time’ and 4 refers to ‘most or all of the time’. Some of the items are: “I think of suicide in order to punish others” and “I feel like dying is better than living like this”. The total scores ranged from 8 to 31 in our sample and they can possibly range from 8 to 32 with higher scores indicating higher suicidal thinking. The SPS scale reported good test re-test reliability and internal consistency (Eskin, 1993; Gençöz and Or, 2006; Atli et al., 2009). In our sample the internal consistency of the subscale was good (α=0.74). The level of acculturation was measured through the Lowlands Acculturation Scale (LAS; Mooren et al., 2001). LAS is a structured questionnaire consisting of 25 statements which represent the difficulties that migrants might face. Some of the items are: “I have frequent contact with (Dutch) people” and “It is important to celebrate the (Turkish) traditional fest in the (Netherlands)”. On the basis of a six-point Likert scale, item scores range from 1 (not applicable) to 6 (very applicable). The instrument is validated among Turkish migrants living in the Netherlands (Mooren et al. 2001). Because of the purpose of the research questions addressed in this paper, we used the adapted version of the scale which was used by Ünlü İnce and colleagues (Ünlü İnce et al., 2014) measuring participation and maintenance as two independent dimensions. These dimensions were created following the two-dimensionality theory of Berry (Beery, 1997) on the basis of the exploratory factor analysis (Ünlü İnce et al., 2014). The 2 subscales are: participation (measured by 4 items) and maintenance (measured by 11 items). The total score for Acculturation and Suicidal Ideation ― 71 participation ranges from 4 to 23, and for the maintenance from 11 to 60 with higher scores indicating a greater degree of participation and maintenance. The new sub scales showed good reliability as chronbach’s alpha was 0.86 for both scales in the previous study (Ünlü İnce et al., 2014). In our sample they also had good internal consistency: participation (α = 0.79) and maintenance (α = 0.80). Hopelessness was measured thorough the Beck Hopelessness Scale (BHS; Beck et al., 1974). BHS is a 20-item true-false inventory which measures a pessimistic outlook for the future. Some of the items are: ‘I don’t expect to get what I really want’ and ‘It is very unlikely that I will get any satisfaction in the future’. Item scores range from 0 to 1 3 and the total scores can range from 0 to 20 with higher scores indicating a greater degree of hopelessness. The Turkish and Dutch versions of the scale reported good internal consistency and test–retest reliability (Seber et. al., 1993; Brown, 2011). The internal consistency of the scale was also good in our data (α=0.74). Attachment styles were measured through the Attachment style Relationship Questionnaire (RQ; Bartholomew & Horowitz, 1991). RQ consists of 4 items each rated on a scale 1 (strongly disagree) to 7 (strongly agree) with higher scores indicating higher characteristics of the particular attachment style for each item on the questionnaire. The 4 items are: secure (comfortable with intimacy and autonomy), dismissive (counter- dependent), preoccupied and fearful (socially avoidant). For example, the secure attachment item is: “It is easy for me to become emotionally close to others, I am comfortable depending on them and having them depend on me, I don’t worry about being alone or having others not accept me” and the dismissive attachment item is: “I am comfortable without close emotional relationships, it is very important to me to feel independent and self-sufficient, and I prefer not to depend on others or have others depended on me”. The Turkish and Dutch versions reported good test-retest reliability and good concurrent validity with the Relationship Scales Questionnaire (Sümer & Güngör, 1999; Griffin & Bartholomew, 1994). 72 ― Chapter 3

Data analysis Descriptive statistics were used to report a gender difference in demographical variables. General Linear Model analysis was conducted to test if two acculturation dimensions, attachment styles and hopelessness were uniquely related with suicidal ideation. The Statistical Package for the Social Sciences (SPSS) version 21 was used for these analyses. Mediation tests between hopelessness, acculturation and suicidal ideation were performed. Interaction between secure attachment, acculturation and suicidal ideation was tested via PROCESS model (Hayes, 2013) using 5000 bootstrapped samples. resUlts Means and standard deviations of study variables are provided in table 1. There were no differences between males and females on any of the study variables. The total mean score for suicidal ideation was low (M=12.64, SD= 3.49) suggesting low suicidal ideation scores among participants in this study. The total mean score on maintenance (M=42.80, SD=8.74) was higher compared to the total mean score on participation (M=9.41, SD=5.02), indicating that the cultural maintenance (i.e. the wish to linkage to those of one’s cultural background) was possibly more common in the current sample. The total mean score for hopelessness (M=9.66, SD=1.66) indicates moderate degree of hopelessness (i.e. a decrease in positive future expectancies). The model accounted for 28% of the variance in participants` suicidal ideation scores. Both acculturation strategies: higher participation (B=-0.18, SE=0.045, p<.000, 95% CI [-0.271, -0.095]) and higher maintenance (B=-0.07, SE=0.03, p=0.031, 95% CI [-0.134, -0.007]) were uniquely associated with lower suicidal ideation. Higher hopelessness was associated with higher suicidal ideation (B=0.03, SE=0.009, p<.000, 95% CI [0.016, 0.050]). Higher secure attachment was associated with lower suicidal ideation (B=-0.03, SE=0.015, p=0.024, 95% CI [-0.062, -0.004]) (see table 2). Acculturation and Suicidal Ideation ― 73

Table 1. Table of mean and standard deviation of study variables (N=185) Variables Total Gender (N=185) Female (N=114) Male (N=71) Mean Standard Mean Standard Mean Standard Deviation Deviation Deviation Age 34.00 11.90 33.41 12.34 34.00 11.30 Suicidal Ideation 12.64 3.49 12.53 3.35 12.81 3.72 Level of acculturation Participation 9.41 5.02 9.84 5.13 8.72 4.81 Maintenance 42.80 8.74 42.78 9.22 42.83 8.74 3 Hopelessness 9.66 1.66 9.57 1.69 9.80 1.60 Attachment style Secure 5.27 1.72 5.21 1.71 5.37 1.73 Dismissive-Avoidant 3.07 1.93 3.16 1.91 2.91 1.96 Pre-occupied 3.86 1.65 3.89 1.63 3.80 1.70 Fearful 3.13 1.83 3.05 1.83 3.25 1.82 The unique associations between acculturation strategies, attachment styles and hopelessness with suicidal ideation after controlling age and gender were tested with the General Linear Model approach. The results of the test are demonstrated in table 2. 74 ― Chapter 3

Table 2. Summary of the general linear model for the study variables associated with suicidal ideation (N=185)

Variables ΔR2 Β (95%CI) SE p

Adj. R2 0.28 Age -0.00 (-0.006, 0.002) 0.00 0.333 Gender -0.04 (-0.042, 0.137) 0.04 0.293 (Male=1, Female=2)

Participation -0.18 (-0.271, -0.095) 0.04 0.000 Maintenance -0.07 (-0.134, -0.007) 0.03 0.031 Hopelessness 0.03 (0.016, 0.050) 0.00 0.000 Secure -0.03 (-0.062, -0.004) 0.01 0.024 Dismissive -0.00 (-0.029, 0.023) 0.01 0.824 Preoccupied 0.00 (-0.024, 0.032) 0.01 0.775 Fearful 0.02 (-0.004, 0.050) 0.01 0.100 Note. The values are unstandardized regression coefficients (B), their standard errors (SE), and 95% confidence intervals (CI).

Indirect Associations between Acculturation, Hopelessness and Suicidal Ideation We have done two separate mediation analyses. For the first, participation was entered as an independent variable; hopelessness was a mediator and suicidal ideation as an outcome. For the second, maintenance was entered as an independent variable, hopelessness as a mediator and suicidal ideation as an outcome. First analysis revealed that, higher participation was significantly associated with lower hopelessness (B = -1.92, SE = 0.34, p< 0.001, 95% CI [-2.58, -1.26]) and higher suicidal ideation (B = 0.03, SE = 0.01, p< 0.001, 95% CI [0.02, 0.05]) (See Figure 1a). The results indicated a significant indirect association between participation and suicidal ideation, (B = -0.07, Boot SE = 0.03, 95% CI [-0.14, -0.02]). In the second analysis, higher maintenance was significantly associated with higher hopelessness, (B = 0.59, SE = 0.25, p = .0203, 95% CI [0.09, 1.09]) (See Figure 1b). Hopelessness was significantly associated with suicidal ideation after controlling maintenance, (B = 0.05, SE = 0.01, p< .001, 95% CI [0.03, 0.06]). The indirect Acculturation and Suicidal Ideation ― 75 association between maintenance and suicidal ideation was also significant, (B = 0.03, Boot SE = 0.02, p < 0.001, 95% CI [0.01, 0.07]) indicating that higher maintenance was associated with higher hopelessness and higher suicidal ideation. Figure 1a. The path analysis indicated a significant indirect association between participation and suicidal ideation (B = -0.07, Boot SE = 0.03, 95% CI [-0.14, -0.02]) through hopelessness. The presented values on the Figure are unstandardized regression coefficients and their standard errors *p<0.001 Figure 1b. The path analysis showing an indirect association between maintenance and suicidal ideation (B = 0.03, Boot SE = 0.02, p< 0.001, 95% CI [0.01, 0.07]) through 3 hopelessness. The presented values on the Figure are unstandardized regression coefficients and their standard errors *p<0.001 76 ― Chapter 3 The path analysis indicated a significant indirect association between participation and suicidal ideation = -0.07, (B Boot = 0.03, SE 95% [-0.14, CI -0.02]) through . 1 c standsc the for total association between independent variable and outcome, whereas stands c’ the for direct association between them the (i.e., association between outcomeindependent after controllingvariable and the effect accountedfor by mediator).

hopelessness. The presented values on the Figure are unstandardized regression coefficients and their standard errors*p<0.001 Figure 1a 1 Acculturation and Suicidal Ideation ― 77

3 The path analysis showing an indirect association between maintenance and suicidal ideation = 0.03, (B Boot = 0.02, SE p< 0.001, 0.07]) 95% [0.01, through CI Figure 1b. hopelessness. The presented values on the Figure are unstandardized regression coefficients and their standard errors*p<0.001 78 ― Chapter 3

Moderated Associations between Secure Attachment Acculturation and Suicidal Ideation We performed two separate moderation analyses. For the first analysis, we have entered participation as an independent variable, secure attachment as a moderator and suicidal ideation as an outcome. For the second analysis, maintenance was entered as an independent variable; secure attachment as a moderator and suicidal ideation as an outcome. In the first analysis, the interaction of participation in the host culture with secure attachment was found to be significant, (B = 0.08, SE = 0.02, p = 0.0002, 95% CI [0.4, 0.12]). As depicted on Figure 2, the association between higher participation and lower suicidal ideation was stronger particularly among those with less secure attachment (B = -0.32, SE = 0.05, p < 0.001, 95% CI [-0.42, -0.22]). The association between higher participation and lower suicidal ideation was weaker for people with more secure attachment (B = -0.05, SE = 0.05, p = 0.3039, 95% CI [-0.16, 0.05]). In the second analysis, the interaction between maintenance and secure attachment was non-significant (B = -0.01, SE = 0.02, p = 0.3449, 95% CI [-0.05, 0.02]). Thus, secure attachment was an effect modifier only for the association between participation and suicidal ideation indicating that among those with less secure attachment, the more they participate in the host culture, the less they think of suicide Acculturation and Suicidal Ideation ― 79

3

Figure 2. The interaction between participation and suicidal ideation, indicating that the association between higher participation and lower suicidal ideation was stronger particularly amongst people with less secure at- tachment. Scores of both participation and secure attachment were centred.

discUssion The present study indicates that Turkish migrants who actively participate in the host culture may have a lower risk of developing suicidal thinking. Participation may protect against suicidal thinking, particularly among those with less secure attachment styles. Overall, the protective role of participation is in line with the literature suggesting a pathway linking participation in the host country with more social support and resilience against suicidal behaviours among migrants (e.g. Bhui et al., 2012; Schweiter et al., 2006). The protective role of participation in the host country contradicts with the studies reporting that the strong identification with the heritage culture increases resilience against mental health problems (Pascoe & Richman, 2009). Yet, it is in line with the research suggesting that cultural maintenance increases perceived discrimination and subsequent psychological distress among women of Turkish descent in Germany (Aichberger et al., 2015a). These contradictory findings in the literature highlights two things which might be helpful to understand what makes acculturation protective 80 ― Chapter 3 or a risk enhancement for mental health. First, it is important to know whether the mainstream culture adoption or the cultural continuity is dominant in migrants` lifestyle. Second, the extent of the culture conflict is relevant. The literature suggests that the greater the disparity between the demands of the host culture and the minority culture, the greater the risk for psychological distress and subsequent mental health problems (Bhugra, 2004). In light of the cultural conflict thesis, it could be argued that the present socio-political conditions such as raising political conservativism, nationalism, racism and subsequent discriminatory politics discourage cultural continuity among migrants. Because Turkish migrants have strong ties with their culture of origin (Klok et al., 2017), the perceived pressure to fit in might lead to opposing cultural maintenance to mainstream culture adoption. Since maintaining traditional cultural values are not welcome on a national level, opposing cultural maintenance might precipitate distress in interactions with other groups, hopelessness and ultimately the risk for suicidal ideation and attempts (Brown & Zagefka, 2011; Phalet et al., 2015; Walker et al., 2008). The interaction of attachment style, acculturation and suicidal ideation suggests that, those who are less securely attached might be more vulnerable for suicidal ideation especially if they also have higher orientation to maintain their ethnic culture. It is known that the insecure attachment styles are characterised by the sense of unworthiness (unlovability) and the anticipation of rejection by the others (e.g. Bowlby, 1973). At the psychological level, this would mean that the current socio-political context is more likely to precipitate this anticipation of rejection by the other groups. Arguably, engaging in daily social interactions with the members of other minority and majority communities might provide them positive intergroup contacts which would contradict the anticipation of rejection and reduce the risk for suicidal thinking and attempts in return (Phalet et al., 2015; Sheftall, 2010; van Acker & Vanbeselaere, 2011). One unexpected finding was that, the suicidal ideation scores were low in our sample (M=12.64, SD= 3.49). This could be due to our sample of middle-aged participants. In cross cultural studies, there is an indication that the percentages for suicidal ideation Acculturation and Suicidal Ideation ― 81 are comparable between Turkish adolescents and adolescents from Europe (e.g. Slovak = 36.4%; Turkish = 33.8%) (Eskin et al., 2014). However, the percentages for suicidal ideation are higher among the Turkish migrant adolescents (38.1%) compared to Dutch adolescents (17.9%) in the Netherlands (van Bergen et al., 2008). It maybe that with a younger sample, the suicidal ideation scores could have been higher. Interestingly, the mean scores for hopelessness (M=9.66, SD=1.66) were high compared to Turkish University student population (M=4.22, SD=4.33) (Zeyrek et al., 2009), Turkish general population (M=4.14, SD=3.72) (Eskin, 2017) and Hispanic migrant sample in the US (M=4.30, SD=3.76) (Dueweke et al., 2015). The low mean 3 scores for suicidal ideation in spite of the high hopelessness scores could be attributed to the possible greater cultural maintenance in our study. Having stronger traditional religious values have been linked with negative attitudes towards disclosing suicidal thoughts amongst the Turkish populations (Eskin, 2003; Eskin, 2004). It maybe that in our sample, disclosing suicidal thoughts was less acceptable than disclosing hopelessness about future because of the shame and stigma attached to publicising suicidal thoughts (Eylem et al., 2016). In contrast to the one of the persistent trends in research on suicide amongst the migrant communities showing great variation between men and women, our results did not reveal such differences. Gender and age disparities in suicidal behaviours exist in Turkish populations (van Bergen et al., 2018). A study from Germany on suicide prevention campaign for women of Turkish descent (Scholer-Ocak, 2015) reported that the most common reason for thinking of suicide was the conflict with the informal network (e.g. family, partner) concerning the family “honour” (33.8%) (Scholer-Ocak, 2015). This would mean that the demands for cultural continuity (i.e. maintaining family “honour”) would be greater among Turkish females which would put them in a vulnerable position for stressful acculturation experiences and suicidal behaviours (e.g. van Bergen et al., 2012). Because of the methodological limitations, this risk group in particular might not be represented in the current study. Future studies must investigate 82 ― Chapter 3 whether different pathways to suicidal behaviours exist among Turkish females and whether there are differences between the first and second generation within this group. It must be underlined that our results should be understood in accordance with the Berry’s bi-dimensional framework which suggests that the acculturation processes are essentially the same for all. One argument against this universalist approach suggests that migrants constantly go through the process of social transformation in line with the changing dynamics of labour forces in highly developed countries (Castles, 2010). This would mean that the process of acculturation is multidimensional and cannot be possibly understood as either migrants participate in the host culture or maintain their heritage culture. It can be both-that is, they can participate in both communities and maintain their heritage culture at the same time. (Borges et al., 2009; Hassan, 1995; Cetin 2015; 2017). The methodological implication of the multidimensional nature of the acculturation construct is, measurement difficulties. The difficulty with measuring acculturation restricts our current understanding of how acculturation is linked with the mental health outcomes (Salant & Lauderdale, 2003). Furthermore, the bi-dimensional model has been criticized for ignoring the dynamics of the intergroup processes central to the acculturation experiences of minority groups as well as majorities in the host country (Brown & Zagefka, 2011; Pettigrew & Trop, 2006). The intergroup relationships perspective suggests that the conflicting acculturation orientations between the minority and the majority groups determine the acculturation experiences of the members of the minority group. The policy towards diversity in the host country is another important contextual determinant of the intergroup relationships and subsequent cultural conflict (Castles, 2010). Future research should take the contextual factors into account and should also investigate if the migrants` interaction with the majority culture can be identified as a protective as well as a risk factor for suicidal behaviours from the viewpoint of the intergroup relations theory. Acculturation and Suicidal Ideation ― 83

Strengths and Limitations The current study has various innovations. We have connected attachment styles with Berry`s bi-dimensional acculturation model. Thus, the results indicate that variations in acculturation experiences could be related with attachment styles. This has been previously suggested by previous research (e.g. Hofstra et al., 2005; van Oudenhoven & Hofstra, 2006). Further, the results also provide new insights on how acculturation experiences might hinder social adaptation leaving Turkish migrants more prone to hopelessness for their future, and suicidal thinking. A further advantage of this study is to study an “at risk” population with constricted 3 reachability (Jansen-Kallenberg et al., 2017). In line with the previous research, it appears that, using multiple channels such as social media and newspapers advertisements in multiple languages are more effective in reaching hard-to-reach populations (Aichberger et al., 2013; Jansen-Kallenberg et al., 2017; Ünlü İnce et al., 2014). There are several methodological limitations to this study. First of all, the current study is cross-sectional which implies that the findings in this study are only a snapshot of reality. Future studies should utilize prospective designs in order to test a broader hypothesis suggesting a path in which higher participation in the Dutch society might be associated with lower suicidal ideation amongst those who are less securely attached. Second, the small number of participants with high suicidal ideation scores resulted in low power to detect possible associations and differences. In spite of this limitation, we found strong associations between hopelessness, participation and suicidal ideation in line with the earlier research. Third, participants were recruited based on a convenience sample. Thus, the sample characteristics might not represent the Turkish migrant population in the Netherlands. It maybe that with a bigger and a representative sample of the Turkish migrant population, higher rates of suicidal ideation and gender differences can be found. Fourth, the self-selection bias also limits the representativeness of the current sample. Future studies could benefit from adjusting for selection bias in their data analysis 84 ― Chapter 3

(Keeble et al., 2015). Another method is to predict the amount of selection bias by using information from non-respondents (Keeble et al., 2015). Fifth, only a limited number of risk factors have been measured. Other factors such as socio economic background (i.e. poverty, unemployment) and education level have to be identified to study the association between demographic risk factors such as gender and suicide outcomes (e.g. Bhugra et al., 1999a,1999b,1999c). Future studies should measure demographic characteristics in a bigger sample. Sixth, the cross-cultural validity of the two new sub-scales: participation and maintenance of the LAS has not been tested. The new sub-scales were created previously by Ünlü İnce and colleagues and showed good reliability (e.g. Chronbach’s alpha was 0.86 for both scales) (Ünlü İnce et al., 2014). In our sample, they also showed good reliability (e.g. Chronbach’s alpha was 0.79 for participation and 0.80 for the maintenance). Further research investigating cross-cultural validity of the new sub-scales in different populations is needed. Another notable limitation was the use of a brief self-report to measure attachment styles in the study. In depth structured clinical interviews might give more reliable results compared to self-report measures. Lastly, we did not study whether the associations between acculturation and suicidal ideation were different between 1st, 2nd and 3rd generation. Different acculturation experiences across generations might be differentially related with the risk factors leading to suicidal ideation (Cetin, 2015; 2017; Klok et al., 2017). Future studies should investigate the differences in the associations between acculturation and suicidal behaviours among first and later generations.

Implications and Conclusions The findings of the current study support the view of healthy culture as the one which adapts to assist people and groups to cope and do better in the world and as such should not be overly rigid or fixed. Participation in the host culture can create opportunities for interactions between groups which would assist group members to adapt to each other. Acculturation and Suicidal Ideation ― 85

Thus, participation in the host country should be an essential element of national mental health policies. On the part of the psychological service providers, an important clinical implication might be including acculturation as a component in screening and treatment for mental health problems (e.g. depression) and suicidal behaviours. The assessment of acculturation and attachment styles may be useful in identifying high risk profiles of migrants. Awareness of mental health professionals of the acculturation experiences as risk or protective factors for suicidal behaviours should be promoted. This approach might help service providers` engagement in service users` help-seeking process. This is central to improving the provision of psychological services for the Turkish populations 3 as well as other migrant groups. Thus, more attention and prominence should be given to acculturation in migrants in general, and further understanding on how it leads to suicidal behaviours specifically.

4

canina kiymak “crUshing life energy”: a QUalitative stUdy on lay and professional Understandings of sUicide and help-seeking among tUrkish migrants in the Uk and in the netherlands

pUblished as Eylem, O., van Bergen, D.D.,Rathod, S., van Straten, A.,Bhui, K., Kerkhof, A.J.F.M (2016). Canına Kıymak ‘crushing life energy’: A qualitative study on lay and professional understandings of suicide and help-seeking among Turkish migrants in the UK and in the Netherlands. International Journal of Culture and Mental Health, 9,1-16. 88 ― Chapter 4

abstract

Background Currently little is known about the views that Turkish migrants hold towards suicide, which may differ from the narratives held by native inhabitants of their host countries. Central to improving the provision of mental health services, furthering our knowledge of these views is important. The aim of this research was to explore Turkish cultural understandings on suicide and help-seeking for suicide.

Methods A qualitative study included data from 6 focus groups (n=4) and 7 individual interviews with 38 Turkish-speaking lay people and 4 key informants living in the Netherlands or UK during the year 2014/15.

Results Through the analysis of participants` stories and narratives, the following key themes emerged in relation to suicide: suicide as an escape from failure and as a failure in itself; acculturation orientation; parenting style and shame and stigma. There were more similarities than differences between the themes among lay participants and key informants from two countries. ‘Canına kıymak’ (crushing life energy) was as a strong metaphor about personal distress. Suicide was perceived as a failing of responsibilities towards the family and community.

Conclusions Future research should aim to give voice to all sub-groups to further the present understanding of suicide and help-seeking processes in these communities. Canına Kıymak ― 89

4 90 ― Chapter 4

introdUction

pidemiological differences in suicide rates between countries as well as within countries have been explained by the context of culture and ethnicity (e.g. Vijayakumar, John, Pirkis & Whiteford, 2005 ; Bhui, 2010a; Hjelmeland, 2011; Lester, 2012). The current suicide research literature is Elimited by the theories generated from research among Western European societies that were treated as relatively homogenous in terms of ethnicity and religion (e.g. Hjelmand, 2011). This limitation brings several challenges to the health care systems especially in improving the provision and accessibility of adequate mental health services for migrants and ethnic minorities. Given these challenges, for instance in the UK, the recent publication of the up-dated NICE schizophrenia guidelines (Updated NICE, 2014) suggests the need for improving the existing knowledge on cultural views of mental health, suicide and help-seeking processes among these highly diverse populations. Currently there is insufficient information on these cultural views of suicide among migrants. The existing literature suggests that besides acculturation orientation, individual meanings and religious beliefs are important factors determining migrant views on suicide (Bhui, 2010a). In the UK, for instance, Muslim communities seem to be more often morally opposed to suicide compared to Hindu communities (Kamal & Lowental, 2002). Several methodological issues exists with these studies. Very often the cultural groups studied are too broad (e.g. Muslims and Hindus) and the cultural, ethnic and religious variations within these broad categories are ignored (Colucci & San Too, 2015). As a result, not much is known about suicide and help-seeking patterns among some ethnic groups within these broad categories, such as the Turkish speaking communities, despite their long history of migration in Western Europe. The Turkish diaspora in Europe mainly comprises three main groups- Turkish Cypriots, mainland Turks and Kurdish communities. Their migration to Europe was often followed by different historical and political reasons (Enneli et al., 2005). Besides the cultural, ethnic and religious differences between these groups, there are Canına Kıymak ― 91 commonalities shared by all as a result of their history of interaction. For example, traditionalism (the strong religious and ethnic identification with their heritage culture) is common among the Turkish migrants in Germany and in Belgium (e.g. Ersanilli & Koopmans, 2010). The traditional family structure emphasizing conformity and the importance of family “honour” seems to persist in Turkish migrant families especially in Germany, in Belgium, and in The Netherlands (Güngör, 2008). In recent years, there is a shift from traditional conservative gender role attitudes toward more egalitarian ones among young Turkish women (e.g. willingness to study, participate in public life, share responsibility) even though their attitudes toward family obligations and parental authority do not often differ from men (e.g. Phalet & Schonpflug, 2011). Suicide is often proposed to be prohibited by the Islam (the dominant religion in this 4 group), and therefore expected to be protective against suicide (Lester, 2006). It is also believed that these populations are protected from suicide because of their traditionalism and strong sense of community cohesiveness (Lester, 2006) as such characteristics would give them access to their informal network (family, friends) to seek help. Nevertheless, the growing literature on suicide in this population indicates that Turkish-speaking communities in Europe face special risks of suicidal behaviours, which reflect the risks in their countries of origin (Schouler-Ocak, 2015). For instance, the age standardized suicide rate in Turkey is 7.9/100000, which is higher than the rates in Iran, Italy, China and the UK (World Health Organisation, 2012). Similarly in the Netherlands, Turkish migrants most often die by suicide at a younger age compared to indigenous Dutch people (Turkish men: 32, Turkish women: 32; Dutch men: 48,Dutch women: 51; Garssen et al., 2006). Furthermore, although completed suicide rates appear to be mostly lower in Turkish migrants in Europe, rates of suicidal thinking and attempted suicide on the other hand do not appear to be lower (e.g. The Netherlands, Belgium, Germany) compared to indigenous populations (Lindert et al., 2008; Burger et al., 2009; van Bergen et al., 2010). The consistency in these suicide trends between mainland Turkey (Sayıl & Devrimci- Özgüven, 2002; Bagli & Sever, 2003) and host countries in Europe suggests the continuity of an interplay between culture specific risk factors (e.g. “honour-related” issues) and 92 ― Chapter 4 universal risk factors (e.g. unemployment) leading to suicide through (or regardless of) the migration process (van Bergen, Montesinos & Schouler-Ocak, 2015, p. 174). The current study explores the cultural views on suicide and help-seeking for suicidal behaviours which are vital to guide the suicide prevention and management efforts in Turkish communities. Given the diversity within Turkish populations and the difficulty to identify a representative Turkish immigrant sample in previous studies (e.g. Ünlü et al., 2013), the current research attempted to include various ethnic, and religious groups representing the Turkish diaspora (i.e. Kurdish, Turkish Cypriot, Alevi and Sunni Muslims) from different generations (first, second and third) in two host countries in Europe. In the existing literature, main limitations for the generalizability of previous qualitative studies, showing the importance of culturally specific factors leading to higher attempted suicide rates among Turkish migrants, are their exclusive focus on one Turkish subpopulation (Alevi, e.g. Cetin, 2015) and the lack of information of Turkish subpopulations participating in the study (Razum et al., 1998; van Bergen et el., 2010). The reason for running the current study in two countries was to see whether differences in historical and ideological contexts preceding migration to these countries and/or host country factors had an influence on Turkish populations` understanding of suicide and help-seeking for suicide. method

Design This qualitative study was held in the Netherlands and in the UK. It concerned in-depth one-to-one interviews with key informants and focus group interviews with young adults from the community aged 18 and above, who had knowledge of suicide events in Turkish communities (through word of mouth or through media), who were either born in Turkey or who had at least one parent and/or grant parent born in Turkey, and sufficient command of English, Turkish or Dutch language. There were no exclusion criteria. Key informants were professionals who were well-known to the community Canına Kıymak ― 93 through their profession or networks. Focus groups were chosen as a method to interview lay people as this method would more likely permit access to a range of perspectives identified within Turkish culture in comparison to individual interviews. Individual interviews with key informants were chosen as a method in order to have more in-depth information on suicide events that key informants come across among in their daily practice. Ethical approval for this study was granted by the Medical Ethical Committee of the VU Amsterdam University in the Netherlands (VUMC) and by the Queen Marry University of London Research Ethics Committee (QMREC) in the UK.

Recruitment and sampling Participants were recruited through social media and through liaison with key people and 4 community groups. A search for relevant non-governmental organisations, community groups or individuals on Facebook resulted in 30 relevant pages, 36 groups and 78 people. A standard text was shared as a post in all these pages and groups. This text was shared in LinkedIn, Twitter and through What`s up application of the i-phone. Lastly, the first author (O.E.) approached potential participants from their networks through face-to-face contact and through e-mail.

Group and individual interview procedures At the beginning of each group interview, participants were asked an open-ended question:

“We are here today to talk about your opinions and knowledge on suicide in the Turkish-speaking

community. Let’s start with what the concept ‘suicide’ brings to your mind and what people try to

commit suicide in your opinion and why?’’

A topic guide was then used to guide the conversation towards the main areas of interest: definitions and views of suicide, knowledge and perceptions of risk factors, normative evaluation of the behaviour, coping behaviour, help-seeking behaviour and barriers, 94 ― Chapter 4 support and views of online help. This guide was established on the basis of relevant literature and discussions with clinicians working with Turkish populations in each country (Netherlands and the UK). Participants were asked to relate their understanding of the Turkish culture to suicide and related help seeking The interviews lasted one hour on average and were conducted at the VU Amsterdam University and the Queen Marry University of London tutorial rooms and in the rooms located in the buildings of the community organisations. All interviews were audio recorded with the informed consent of the participants. All participants consented to publication of their interview data using pseudonyms.

Analysis Interview recordings were transcribed verbatim. Pseudonyms were assigned to each interviewee. Thematic analysis was conducted that aimed to identify, analyse and report themes within the participants` and key informant`s descriptions of their experience of suicide, seeking help and disclosing suicidal behaviours referring to their understanding and experience of the Turkish culture (Braun & Clarke, 2006). First, the code system (and categories and themes developed on the basis of the coding process) was developed gradually and collaboratively. The code system developed on theoretical grounds and included following categories: Definition of suicide, normative evaluation of suicide, risk factors leading to suicide, protective factors against suicide and normative evaluation of help-seeking for suicide. Each of these categories had a number of sub-categories and codes. Most important categories were: Normative evaluation of suicide, risk factors leading to suicide and normative evaluation of help-seeking for suicide. The definition and protective factors were less central compared to the others and helped interpreting the information coded under these main categories. This coding system was developed by O.E (first author) and was checked independently by F.J.I and J.I. Once an agreement was reached, they were further developed, refined and applied to the transcripts. O.E. was the main coder and J.I involved as a second coder, who systematically counter- checked the coding, to assure the robustness and internal validity of the coding. The Canına Kıymak ― 95 data was coded manually. Disagreements over the coding were discussed between the main and the second coder and where necessary external experts were consulted. Detailed descriptive accounts were produced for each major theme alongside the related extracts from participants’ transcripts. Analysis continued until no new themes emerged from the transcripts. The following themes were validated between the coders and considered more central than the other themes as they were central to the discussions in relation to culturally specific issues related with suicide in Turkish populations. These were: Suicide as an escape from failure, and as a failure in itself, acculturation orientation, parenting style, shame and stigma regarding suicidality and help-seeking (see table 2).

4 resUlts

Overview of the data collection process and notes about emotions at the meeting Forty two people of Turkish descent, aged 18-63, 35 women and 7 men participated in the study. Table 1 shows the demographic characteristics of participants in two countries. During the data collection process, one of the challenges was to identify ethnocultural background of the participants. Most of the time, ethnocultural composition of focus group meetings was mixed. Thus, it was decided to let participants self-define themselves with any ethnocultural membership. In the UK, 10 participants identified themselves as Sunni Muslim and traditional (strong religious and ethnic identification with their heritage culture) and 2 participants identified themselves as Kurdish Alevi and secular. Only 2 of them did not identify themselves with any ethnic background and religion. In the Netherlands, 9 participants define themselves as Sunni Muslim and traditional, 2 of them considered themselves as secular Arabic Alevis and 1 did not identify themselves with any ethnic background and religion. 96 ― Chapter 4

Table 1. Demographic characteristics of participants Individual interviews Individual interviews Focus Group Interviews with Key Informants with lay people with lay people The The The UK Netherlands The UK The UK Netherlands N=2 N=2 N=3 N=14 N=12 Gender n (%) n (%) n (%) n(%) n (%) Male 1(50) 2(100) 1(33) 3 (21) 1 (8) Female 1(50) 0 2(66) 11(78) 11(91) Age 18-35 0 1(50) 0 6 (43) 8 (66) 36-49 1(50) 0 1(33) 3 (21) 2 (16) > 50 1(50) 1(50) 2(66) 5 (36) 2 (16) Profession Employed 2(100) 2(100) 0 4 (28) 4 (33) Unemployed 0 0 3(100) 10(71) 8 (66) Education level None/ primary 2(100) 0 2(66) 6 (42) 0 Middle 1(50) 0 1(33) 2 (14) 1(8) Higher 2(100) 2(100) 0 6 (43) 11(91) education Note. Number of lay people from the1stgeneration:11, 2ndgeneration: 4and 3rd generation: 6 in focus group interviews in the UK Number of lay people from the 1stgeneration:2, 2ndgeneration: 4 and 3rd generation: 8 in focus group interviews in the Netherlands

Another challenge was about the discomfort associated with discussing suicide in a group. In the UK, some participants did not want to be in a group setting. O.E. interviewed these participants alone. Two of them identified themselves as Turkish Cypriot and had close relatives who had history of suicide. More similarities than differences were observed between the narratives of key informants and lay people across the countries. Thus it was decided to analyze the key Canına Kıymak ― 97 informant and lay people transcripts together. The few differences, that were observed, are noted and discussed in the results and discussion section. Lay people named several metaphors related with suicide and help-seeking (see table 3). “Canına kıymak” (crushing life energy) was the strongest metaphor which was associated with distress and suicide and was used commonly by many first, second and third generation participants in both countries. Through the analysis of all the narratives, seven themes appeared in relation to participants` attitudes and conceptualizations of suicide, coping and help-seeking patterns in Turkish populations. Four of these themes: suicide as an escape from failure and as a failure in itself; acculturation orientation; parenting style and shame and stigma regarding suicidality and help-seeking are presented and discussed in detail. Table 2 4 summarises all themes that emerged and shows the variation across countries, different participants and generations. 98 ― Chapter 4

Table 2. Themes related to perceived causes of suicide, emerging from focus group and individual interviews with 42 people (35 women, 7 men) of Turkish descent, and aged 18-63, in the Netherlands and in the UK The Netherlands The UK Themes related to perceived causes of suicide N=16 N=26 N (%) N (%) 1. Suicide as an escape from failure, and failure in itself 9 (56) 13 (88) 2. Suicide is an act of overt and covert aggression 5 (31) 6 (26) 3. Acculturation orientation 15 (94) 7 (27) 4. Parenting style 8 (50) 9 (35) 5. Shame, and stigma regarding suicidality and help-seeking 16 (100) 13 (50) 6. Coping and somatization 7 (44) 11 (69) 7. Perceived intrusiveness and a disapproving attitude in a helping 10 (62) 4 (15) relationship 1. Mostly among all traditional participants from (1st, 2nd and 3rd) generations in both countries 2. Mostly among 1st and 2nd generation Kurdish, Turkish Cypriot participants and among key informants in both countries 3. Mostly among 3rd generation traditional participants in the Netherlands and 1st and 2nd generation Alevis and key informants in both countries 4. Mostly among all 1st generation participants in both countries 5. Mostly among all participants and all generations in both countries 6. Mostly among 1st generation traditional participants and key informants in both countries 7. Mostly among traditional participants from all generations in the Netherlands

1. Suicide as an escape from failure, and as a failure in itself This theme was especially relevant to strongly traditional participants in the Netherlands and in the UK. Alevi, Kurdish participants and key informants with secular ideology often did not speak about suicide as a failure. Many traditional participants referred to suicide as an act to escape from the consciousness of failure in meeting religious objectives for being a good person. Some lay participants used metaphors such as “şeytana uymak” and “imtihanı kaybetmek” (see table 3) often when religion was mentioned in relation to suicide. Canına Kıymak ― 99

Some of them referred to life events (such as: loss of a loved one, break up) as “exams” from Allah to test the capacity to endure pain. According to this perspective, suicide action is a failure in itself and meant lacking religious values: patience, endurance and self-sacrifice which were important to pass the life exam in order to deserve a good afterlife:

“According to my faith (Sunni Muslim), death is only the beginning. It is the beginning of the real

life and what we are experiencing now is only an exam. Committing suicide means failing that exam.

The pain God has given you is exactly what you can tolerate” lay participant, the Netherlands

Several participants believed that culture bound gender roles (i.e. family leader role of 4 man; subordinate role of woman who is required to conform) were embedded within the Islamic faith. They argued that failing to fulfil these gender role expectations was often followed by gossip in the community and led to isolation of the person who had failed. They spoke about gossip as the worst thing that could happen to someone. Under these circumstances, suicide was often a solution to escape from the idea of failing and associated feelings of shame. As part of these gender roles, namus (honour) was often defined as a collective concept showing the family`s standing in the community. Some participants used the metaphor: “his man`s pride was hurt” (see table 3, metaphor 3) in order to describe a situation in which a man felt ashamed for failing to practice his manhood which meant failing to protect the “honour” of the family:

“I know a man. He found out that his wife was cheating on him. This was followed by gossip

in the community. His “man’s pride” was hurt. He couldn’t tolerate the gossip any longer and he

committed suicide” lay participant, the Netherlands 100 ― Chapter 4

“We usually hear young girls committing suicide to protect the “honour” of the family. If they

lose their virginity before they get married, the society finds their suicide attempt normal” lay

participant, the Netherlands

As the preceding quotes indicate, for several traditional participants, culture was an important context teaching people to live for the community instead of themselves. Some discussed an image of a person who had lived in divergence with the norm of their systems that were thought to protect the person`s well-being, such as religion, family and community. It was indicated that this isolated state was justified especially when gender role expectations were not fulfilled. In these circumstances suicide was normalized in the community.

2. Acculturation orientation This theme consistently emerged between various Alevi, Kurdish and traditional participants in two countries. Many referred to diasporic culture as a “separate culture” comprising traditional values from their mother culture as well as from Dutch and British values they had acquired during their settlement process. Many third generation participants spoke about a struggle with their choice of an acculturation strategy: whether they should maintain their heritage culture or they should integrate into Dutch/ British culture. Variation in participants` views about how this struggle was related with distress and suicide emerged. Some traditional lay participants mentioned this struggle as an advantage encouraging them to take responsibilities (e.g. learning how to manage their personal workload whilst helping their relatives) from an early age. They argued that failing to acknowledge differences between cultures could result in separation from their protective systems (family, community) and suicide. By contrast, several key informants and Alevi participants believed that this struggle was in fact a disadvantage as older generation’s expectations from the younger generation to choose heritage culture often exacerbate the distress which sometimes leads to suicide: Canına Kıymak ― 101

“You are expected to choose a group. In my experience 90% of them choose Turkish culture, hence

their parents. When it comes to choosing... I find it very difficult. Why do you have to choose? Once

you choose then you have less space for personal development. I never identified myself with being

Turkish. That made it easy to improve myself in terms of religion and culture. Once you don’t have

this comfort, I think you are more likely to experience conflicts with your friends and with your

family. Eventually if there is no development and no communication then psychological problems

start and suicide is part of this picture” lay person, the Netherlands

Several key informants believed that a perceived pressure to choose heritage culture often led young people unconsciously or consciously to split between different roles they play in different occasions. Some argued that although this strategy seemed adaptive to be 4 able to fit in to different contexts (school, family, and neighbourhood), it was perhaps mentally exhaustive for younger generations which precipitated further conflicts and distress related with their uncertainty about their identity:

“Young people try to integrate and spend every single day fighting for it... They play a different

role at home with their parents, a different role at school and a different one with their peers in

their neighbourhood. This exhausts them as far as I can see. They play three theatres every single day”

key informant, the UK

For some lay participants, “splitting” followed a trajectory leading to suicide. They believed that splitting made young men vulnerable for participating in gangs and using drugs. They argued that although this might have seemed as a way of boosting their confidence as a man, being a criminal and police arrests most often brought shame to the family. Their resentment and regret sometimes gradually separated them from family and community. For these young men, suicide might be an escape from this marginalized position:

“ You know gang culture exists... Suicide is common among people belonging to those gangs. Once

they are asked whether they are coward or chicken by their peers then they are provoked to prove 102 ― Chapter 4

the opposite and they become a member of a gang... Then mafias take them in. People who want

to escape, end up committing suicide” key informant, the UK

Several participants argued that a similar trajectory leading to a marginalization and suicide is also possible for young women. Many young traditional female participants often expressed a difficulty with their perceived pressure on prioritizing social obligations whilst fulfilling their school-related or work-related obligations. Some key informants believed that imposing gender roles was much stronger on women in a traditional family structure. They believed that young women, who are coming from this strict religious background, were more likely to question their parents’ understanding of gender roles as these values were in conflict with the idea of gender equality that they learned from their experiences with the host culture. This was identified as a distressing factor often encouraging women to move out of the family home, leading to intergenerational conflicts, resentment and suicide attempts:

“This is especially difficult for girls... I knew a family... Once the mother said ‘my daughter washes

her brother’s feet’ in front of her daughter. The girl left the room immediately... Then I asked her

why does she have to do that? She said ‘she is a girl of course she has to do it’... After a while the girl

became very rebellious and moved out. Her father constantly threatens to kill her without even

questioning themselves” key informant, the UK

3. Parenting style Variations in opinions about the role of parenting in suicide events emerged between old and younger generations from different backgrounds in two countries. First generation traditional lay participants spoke about permissive parenting (not being aware of children’s whereabouts) as the main reason for young person’s involvement with gangs and drugs. Many argued that parental control (i.e. questioning young people and being aware of their whereabouts) would protect them from getting involved in gangs which may result in suicide attempts. Canına Kıymak ― 103

Several second and third generation lay participants however, did not see parental control as protection from suicide. By contrast, they identified these controlling attitudes as “over-protective”, encouraging suicidal people to distance themselves even more from their parents. They argued that this would eventually lead to “bunalım” (see table 3, metaphor 5) and suicide. Some traditional lay people believed that over-protective parenting style had greater contribution in suicidal events among young women as they were expected to be family oriented (e.g. spending more time with the family). They further argued that men could escape from “being over-involved with the family” to some extent for example, through their sport activities, whilst women did not. Some key informants spoke about a breakdown in the traditional family structure as a result of the migration process. They argued that young generations gained status 4 in a family structure as a result of their language proficiency (in Dutch or in English). In contrast to the traditional family structure emphasizing parental control, parents started to negotiate with them as children grew up. According to these key informants, repetition of negotiations often resulted in a type of relationship between children and parents based on interests, leaving youth ill equipped to function in society:

“Children grow up in a different world until they turn to 6. They experience lack of affection from

their father who is almost absent.... The mother, who is constantly oppressed, shows ‘chicken-type’

protection to her children. Instead of showing affection, she tries to take them under her wings

like a chicken. Then a sort of relationship based on interests starts.... ‘If you don’t do this we will buy

this for you’. Then children, growing up in this world, believe that life is based on this give and take

principle. But once they go to school, they realize that life outside the house is completely different”

key informant, the UK

Furthermore, the sort of relationship based on interests sometimes resulted with parentification (reversed parent and child roles). Accordingly, children who were expected to take responsibilities of a parent (e.g. accompanying a parent during GP visits, translating the personal information related with the presenting problems of a 104 ― Chapter 4 parent) from an early age would develop the belief that they should do everything by themselves. Some key informants referred this belief as an important reason limiting Turkish migrants` problem solving skills such as asking for help and perspective taking. They believed that this destructive thinking style (i.e. overly self-reliant) sometimes put Turkish people in an outsider position such as having difficulty in fitting in school life or sometimes encouraged them to express their emotional problems physically (somatisation) or through a suicide attempt.

4. Shame and stigma regarding suicidality and help-seeking Shame and stigma in relation to suicide disclosure and help-seeking consistently emerged among all participants in two countries. There were few differences between key informants and lay participants about help-seeking. Some key informants argued that help-seeking was delayed until suicidal thoughts and emotional problems piled up and resulted with a crisis situation. For many lay people however, informal help- seeking: talking to relatives and friends within their network was usually seen as an adequate substitute to consulting formal mental health care services. Consulting religious leaders was identified as a common help-seeking strategy only among traditional lay participants who were deeply religious. Many lay participants indicated that people with suicidal behaviours often did not use available formal health care unless it was recommended to them by their social network. Several barriers to help-seeking were mentioned. They consistently spoke about interplay between shame and stigma (i.e. rejecting attitude to suicide disclosure) often leading to reluctance to seek help:

“She (a relative with suicidal thoughts) was only talking to herself, she was keeping everything to

herself. She was so ashamed” lay person, the UK Canına Kıymak ― 105

“You think of others before you talk about your issues in public. You try to cover it up so no one

can hear about it in the community. Because if they hear, they will stop talking to you, they will

think you are crazy” lay person, the Netherlands

Many lay participants and key informants believed that the person with suicidal thoughts often felt ashamed and rejected in response to comments indicating that the suicidal person was well-off materially and it was not understandable why the person was suffering:

“Ah poor you…. Why are you so sad, you have got everything you want in your life- you have a

house, you have money, you are the only son of your family so what else you want from life why 4

are you making big deal out of this?”lay person, the UK

Several participants believed that covering it up and keeping it to oneself were often used to hide emotional ups and downs related with suicidal thoughts. Many lay participants believed that in some instances, relatives or friends were able to realize such changes in suicidal person`s mood or behaviours. Some argued that relatives sometimes tried to occupy the person with light hearted and fun activities instead of talking directly about the changes:

“ They don`t try to seek help. If they want to there is no barrier. They try to solve their (children’s)

problems themselves sometimes through money, sometimes through more protection or sometimes

through occupying them with activities. But these attempts are not solution to their children’s

problems.” lay person, the UK

Several participants believed that when the suicidal thoughts were related with family problems, people often preferred to keep it to themselves One Turkish Cypriot participant used the metaphor “acı söz” (see table 3, metaphor 4) in order to describe the mixed 106 ― Chapter 4 feelings of resentment and hurt when a loved one talks down to someone which could cause suicidal thoughts.

“Sometimes you are hurt by your loved ones (family) the most. But if this is the case, then you

hesitate to talk to others or professionals because you don`t want to betray your family” lay person,

the Netherlands

Several key informants spoke about stigma associated with going to a psychologist or a health professional in the community. In some occasions, several lay participants from third generation attributed this reluctance to the existing belief that if they use health services, it will be a bad reference for them in their future job applications. One key informant spoke extensively about how he experienced stigma associated with seeing a professional and adjusted his services according to Turkish populations in their practice:

“The therapeutic relationship with Turkish service users is based on negotiation. They ask for a

letter. Then I tell them that if they want a letter then they should also open up themselves more.”

key informant, the Netherlands Canına Kıymak ― 107

Table 3. List of metaphors appeared in relation to suicide and help-seeking during the interviews Metaphors Meaning 1. Canına kıymak “Can” stands for life energy and “kıymak” is a process of producing minced meat or cutting objects into small pieces. This is commonly used among lay people to talk about suicide. 2. Şeytana uymak Failing to act according to God`s will which will be punished in life after death. 3. Gururuna yedirememek Taking something personal and turning it into a pride issue. It is often used by men to describe a situation where they feel provoked to prove someone wrong. 4. Acı söz A thoughtless word (bitter taste) making someone feel humiliated 4. İmtihanı kaybetmek In the context of religion, it means failing to pass the life exam in 4 material world hence failing to deserve a good afterlife. 5. Bunalıma girmek A process in which people silently keep everything to themselves. “Bunalım” is typically used to talk about depression among lay people. 6. Çileden çıkmak Losing the grip as a result of “bunalım” 7. Cinnet geçirmek Losing control, not being able to take anymore.

discUssion This qualitative study has identified views, meanings and metaphors related with suicide and help-seeking among Turkish lay people and key informants in the Netherlands and in the UK. The observations noted more similarities than differences within these narratives among key informants and lay people in two countries. The findings indicate that the understanding of suicide as failing of responsibilities towards the family and community is central to the main stressors (acculturation orientation, transformation in family system) leading to suicidal behaviours. In the help-seeking process for suicide, feelings of shame (putting down the family) and the stigma associated with suicide are identified as major barriers limiting Turkish migrants’ access to their informal and formal networks to seek help. The findings further suggest that, within these informal 108 ― Chapter 4 networks, emotions and personal issues related with suicide are not discussed and suicidal people often feel rejected in response to their relatives’ attempts to cover these issues up. One explanation for observing more similarities in two diaspora could be related with the role of social network (e.g. family) regulating and overseeing the beliefs and attitudes about suicide and help-seeking for suicide. Arguably, this shared knowledge of suicide is not so much influenced in their countries of settlement. The most remarkable example for this was the metaphor: “canına kıymak” which was well-known and understandable to all participants from different generations and from different backgrounds. This metaphor refers to self (can) as a collective being which is part of relatives and the whole community. Destroying this self (kıymak) means destroying all the co-existing systems (family, friends) together with the self. Another explanation for the existing similarities between perspectives of participants could be related with little number of participants representing some communities (e.g. Alevi, Kurdish) in the current study. In fact, some differences were observed in the sample. For instance, traditional participants, who identified themselves as deeply religious, used “we” language much more compared to Alevi and Kurdish participants who sometimes did not identify themselves with any community. The religious values connecting an individual to bigger social networks (e.g. family, community) and protecting them from suicide was often considered oppressive by Alevi and Kurdish and participants. Thus, more differences between participants from different backgrounds could have been observed if the sample size was enhanced for each of various groups representing Turkish populations. The link between suicide and young generation’s concerns over failing responsibilities towards the family and community has been observed in many other populations sharing collectivistic cultural values such as Gypsy communities (Lester, 2015), indigenous American community (e.g. Lester, 2012, p.71), Chinese communities (Tseng and Wu, 1985, p.15), African communities (Mugisha et al., 2012) and South Asian Muslim communities in the UK (Till & Bhugra, 2015; Kamal and Lowenthal, 2002). This study suggests that, in Turkish populations such concerns are often precipitated by the intergenerational Canına Kıymak ― 109 conflicts over younger generation’s choice of an acculturation strategy. Compared to young men, introduction of Western values such as autonomy and freedom of expression perhaps causes greater conflicts between young women and their parents during the acculturation process. Arguably in a traditional family structure, the stigma of women who have gone astray (or spoilt the namus) (van Bergen et al., 2010) is more distressing for young women of Turkish descent and more likely to lead to splitting between multiple roles (e.g. modest girl with the family helping with the household chores, an autonomous woman at work). This coping strategy may lead to an ambiguous position in their family and wider ethnic community which may result with suicide in the long run. However, for many young men, forming gangs perhaps does not clash with the traditional gender role values (assertiveness, superiority of men) and does not initiate cultural conflicts with 4 parents in the beginning. Later on, their regret and resentment for being a criminal perhaps gradually separate them from family and community and lead to “bunalım” preceding suicidal behaviours (Cetin, 2015). The findings have also called attention to the shame and stigma associated with suicide which limit suicidal Turkish people’s access to their informal (e.g. relatives and friends) and formal (e.g.GPs) networks. This finding is inconsistent with the cross- cultural studies by Eskin and colleagues suggesting the presence of more accepting and helping reactions to suicide disclosures among Turkish young adults compared to number of European populations (Slovakia, Switzerland, Austria) (Eskin,1999a;1999b; Eskin et al., 2010; Eskin et al., 2015). These contradictory findings could be explained by the individual meanings attributed to the helping reactions in the current study. It appears that within Turkish migrants’ informal networks, relatives` or friends` reactions to suicide disclosure might be perceived as rejection which would exacerbate suicidal thoughts associated with being a failure and feelings of shame. For example, observations about people`s attempt to keep suicidal person occupied with fun activities may look like helping reactions but they may in fact block the communication about real issues causing suicidal thinking. 110 ― Chapter 4

In the current study, the observations about the limited access to informal and formal networks to seek help further suggests the presence of ruminative processes about failure and aversive feelings (e.g. feeling humiliated, feeling ashamed). It could be argued that, ongoing conflicts (e.g. “honour” related issues) between generations elicit negative emotions (such as feeling ashamed, feeling guilty or feeling resentful). These aversive feelings would make it difficult to divert someone`s attention away from these intense feelings through available coping behaviours (e.g. praying). Failing to rely on existing coping behaviours or the lack hereof, would increase the chances of using somatization or self-harm as a coping strategy as physical sensation of pain would seem like an available option to shift the attention away from the ruminative thoughts about the intense aversive feelings (Selby & Joiner, 2009). Observations for thought suppression (e.g. Bunalim: a process of talking to oneself) provide support for this model. Further support for the relevance of this model for Turkish populations comes from a cross-sectional study suggesting that thought suppression has a strong relationship with behavioural dysregulation (i.e. self-harm) among Turkish university students in Turkey (Tuna & Bozo, 2014).

Strengths and Limitations The current research design among various Turkish groups addressed several methodological issues identified as critical for researching suicide in immigrant communities: trust-building, confidentiality and reaching a diverse sample (Colucci & S. Too, 2015). The current research attempted to include participants from various ethnocultural backgrounds within the Turkish populations from different generations in two countries. Bi-lingual student researchers, who were familiar with these communities, were helpful especially in liaising with various community groups. Additionally, O.E. tailored the research approach according to the tangible requests of these community groups. For instance, individual interviews with some lay people, who did not want to be in a group meeting, were organised in order to respect their concern for being labelled (i.e. Canına Kıymak ― 111 being related to a suicidal person) in the community. This tailored approach helped participants to distance themselves from the topic as they were more eager to talk about their experiences despite their initial unease to talk about suicide. There are several limitations with this study. It was difficult to identify participants’ ethnic membership due to the mixed ethnocultural composition of focus groups. Arguably, some participants avoided to self-identify themselves belonging to any ethnic membership in a group setting or it was safer to identify oneself with traditional and Sunni Muslim (representing dominant religion) identity in a group setting. As a result, there were only few people from Kurdish, Alevi and Turkish Cypriot backgrounds. Despite the little number of these participants, some differences between their views on suicide and help-seeking were indeed observed. Thus, future research should aim to give 4 voice to all sub-groups to further the present understanding of suicide and help-seeking processes in these communities. Another limitation was not including lay participants and key informants currently residing in Turkey. The lack of comparison group from mainland Turkey limit the conclusions from the observations about Turkish cultural views on suicide and how these views were maintained or re-constructed in two different diasporas. Future research should include a comparison group from mainland Turkey to enhance the generalizability of such observations. Lastly, the data for the current study were based on retrospective accounts of memories and experiences of professionals and lay people about suicide and help-seeking for suicide. Although participants` willingness to be involved in the research process was observed, their recounting of memories of suicide and help-seeking in their communities were likely to have been influenced by many factors related with circumstances prior to or during the interviews. This study does not claim to represent all the risk factors and mechanisms leading to suicide in all Turkish migrant populations given that the communication between suicidal people and their networks (informal and formal) is often limited. 112 ― Chapter 4

Implications and conclusions To conclude, this research has important implications for research and clinical practice with Turkish populations and possibly, with other socially excluded immigrant communities. The understanding of suicide as a failure in relation to the family and community expectations was found in Turkish populations, and was associated with shame and stigma leading to reluctance to seek-help. The research process supports a tailored research approach, based on trust-building, to engage migrants. Researchers hope to inform professionals about the lay perspectives on suicide and help-seeking process which are vital to tailor the mental health care delivery according to these populations. For instance, instead of directly questioning the presence of risk factors for suicide, which may trigger the feelings of shame, practitioners may find it useful to use metaphors (e.g. bunalım, canına kıymak) to explore the ruminative processes and presence of suicidal thoughts and individual meanings attributed to such thoughts. This approach may help with the rapport building. Lastly, while professionals need to tailor their approach, the informal network (friends and relatives) of suicidal people should be also educated about risk factors, available resources for help and helpful reactions to engage with suicidal people (e.g. referring to professionals) through community-based suicide prevention campaigns. Canına Kıymak ― 113

4

5

attempted sUicide and sUicide of yoUng tUrkish Women in eUrope and tUrkey: a systematic literatUre revieW of characteristics and precipitating factors

sUbmitted as van Bergen, D., Eylem, O., Heredıa-Montesinos, A. (2019). Attempted suicide and suicide of young Turkish women in Europe and Turkey. A systematic literature review of characteristics and precipitating factors. 116 ― Chapter 5

abstract

Background The heightened risk for suicidal behaviour among Turkish women living in Turkey and Europe is a serious public health problem. To date, there are no systematic reviews investigating whether demographic factors, psychiatric or physical illness, social stressors (e.g. domestic violence) and/or migration related factors are relevant to understand their suicidal behaviours. This study synthesises and compares empirical evidence of precipitating factors and characteristics for suicide and attempted suicide between Turkish women in Europe and Turkey.

Methods We systematically searched PsycINFO, PubMed, Med Line, Web of Science, Smart Cat, Safety Lit, BASE and Ulakbim. Turkish, German and Dutch translations of the key words were used to retrieve the Turkish, German, and Dutch literature. The reference lists of the included papers were also checked for relevant studies. We extracted data on the country/region, population characteristics, sample, recruitment, methods of data collection, type of suicidal behaviour (attempted suicide or suicide) and precipitating factors and characteristics. The results are synthesized qualitatively.

Results We retrieved 8 studies on attempted suicide in Europe, 16 on attempted suicide in Turkey, and 10 studies on suicide in Turkey (34 in total). There were rather similar precipitating factors and characteristics of attempted suicide and suicide in Turkey and in Europe. Social precipitants (i.e. marital conflict, domestic violence and “honour related” violence) were frequently reported across the studies. Attempted Suicide and Suicide ― 117

Conclusion The strong role of the social precipitants points to the possible cultural continuity of patriarchy and oppressive practices accounting the increased risk of suicide and attempted suicide among Turkish females. The framework of intersectionality is relevant since structural inequalities in gender roles and expectations (i.e. pressure to protect ‘honour’) and power imbalances in social class (such as poverty) may lead to increased risk for interpersonal violence and eventual suicidal behaviours among Turkish women. Thus, suicide prevention efforts should address the underlying cultural beliefs and attitudes through education programmes.

5 118 ― Chapter 5

introdUction

t is well documented that women attempt suicide more often than men, while men die by suicide more often than women worldwide (Beautrais, 2006; Hegerl, 2016). An exception to this longstanding pattern is the high number of women who die by suicide compared to men in Muslim-majority communities and Icountries (Karam et al., 2007; 2008; Lester, 2006; Rezaeian, 2010; Canetto; 2015). This variation is seen even more so when stratified by region. For instance, young Pakistani women living in the rural and/or remote area of Pakistan (Khan et al., 2009), women in rural Iran (Aliverdinia & Pridemore, 2009) and Turkish women living in rural areas of Eastern part of Turkey have a higher risk of suicide compared to men (Bağlı & Sever, 2003; Razum & Zeeb, 2004). Regarding Turkish women, studies have shown a similarly disproportionate risk of suicidal behaviour among those who have migrated to European countries, such as Germany, Switzerland or the Netherlands (Schouler-Ocak, 2015). This is why the suicidal behaviours of Turkish women who live in Turkey or Europe need further investigation. Since 2000, the suicide rate among Turkish women aged 15 and 24 in Eastern Turkey is twice as high as their male counterparts (Devrimci-Özgüven & Sayıl, 2003; Bağlı & Sever, 2003). These findings were replicated by more recent annual reports on suicide statistics representing the general population of Turkey (e.g. Öner et al., 2007; Öner et al., 2015). Similarly, hospital registration data in the South East region of Turkey has also shown that attempted suicide rates were 4.15 times higher among women (15–24 years old) who presented at emergency departments compared to men of the same age as well as older men and women (Simsek et al., 2013). In the Netherlands (Burger et al., 2006), Germany (Lizardi et al., 2006) and Switzerland (Brückner et al., 2011), the risk of attempted suicide was at least 2-5 times higher among Turkish migrant women aged 14 and 25 who presented at emergency departments compared to Turkish immigrant men of the same age and women in the host country. Further, girls from Turkish descent aged 10 and 17 had an almost 2 fold Attempted Suicide and Suicide ― 119 elevated risk for suicide compared to the ethnic majority women in Germany (Relative risk=1.79; 95% CI = 1.41-2.27; Razum & Zeeb, 2004). The heightened risk of suicidal behaviours among a sub-group of Turkish women in Turkey and Europe poses two important questions: 1) What factors contribute to the increased risk of suicidal behaviour of Turkish young women? And, 2) to what extent there is continuity (or change) in the factors contributing to their suicidal behaviour when studies conducted in host countries are compared to those in Turkey? One of the well-known important factors contributing to the increased risk for suicidal behaviours is mental illnesses, notably depression, worldwide (O’Connor, 2016). In Turkey, epidemiological studies revealed that the prevalence of depression is ranging from 4.4% to 48% (Binbay et al., 2017; World Health Organisation, 2017). In terms of the gender differences, depression was three times more likely to be reported among female 5 attempters (72%) than among male attempters in Turkey (27%; Saraçoğlu, et al., 2014). Similarly, depression is more often reported among Turkish female suicide attempters (51%) compared to the female suicide attempters from majority ethnic group (33,9%) in Germany (Aicherber et al., 2015). Although Turkish female suicide attempters` higher rates of depression is an important factor, their distress can only be understood within the socio-cultural context (Hjelmeland & Knizek, 2017). Thus, it is necessary to investigate the role of the gender- related factors such as psychosocial stressors and trauma (i.e. domestic violence, sexual and physical abuse, “honour-related” violence) since women are more often victims of these events compared to men (Beautrais, 2006; van Bergen, Eikelenboom & van de Looij-Janssen, 2018; Rezaeian, 2010; Canetto, 2015).

Socio-cultural context Turkish migrant populations in Europe are one the largest ethnic minority populations. The majority lives in Germany (13.52%) and in the Netherlands (2.3%; Kilberg, 2014; Statistisches Bundesamt, 2019). Migration from Turkey to Western Europe has been mainly characterised by two influences: 1) Socio-economic deprivation especially in the 120 ― Chapter 5 rural areas of the Eastern Turkey, and 2) ethnic conflict between the Turkish and Kurdish citizens (Enelli, Modood & Bradley, 2005). Additionally, there is an ongoing political persecution since 1980s and it has been increased in recent years due to the coup attempt in 2016 (Ozdemir et al., 2019). Turkey`s cultural landscape is characterized by a diverse set of values and norms. Traditionalism (i.e. the strong identification with Islamic religion and patriarchal norms and values) is more common in the Central and Eastern regions, while devotion to secular values and egalitarian view of gender roles are more often seen in the West. The concepts of “namus” and “seref” (i.e. honour) are central to the definition of gender roles in patriarchical norms and values (Eylem & Eskin, 2019). In a rural family context, women are responsible for the “honour” of the family by maintaining their sexual abstinence until marriage. Failure to do so results in the family`s loss of reputation among the community. This custom brings strong scrutiny on behaviours of women and leads to personal autonomy restrictions in different areas of life. Some typical examples of autonomy restrictions include child marriage, forcing women to stay in an unwanted marriage, discouraging women from seeking employment (Hekimoğlu et al., 2016; Altındağ, 2005; Heredia Montesinos et al., 2018). In extreme cases, “honour” killing is practiced by the male members of the family and/or community when a perceived dishonourable act of women jeopardizes the family`s reputation in the community (Eylem & Eskin, 2019). “Honour-related” violence (i.e. being accused of not maintaining one`s sexual abstinence, being forced to marry through their parents, pressure from partner or family to stay in an unwanted marriage), aims to protect the family “namus” and it could be one of the precipitating factors for suicidal behaviours among young Turkish migrant women in Germany (Schouler-Ocak, 2015), in the Netherlands (van Bergen et al., 2009) as well as among Turkish women in rural areas in Turkey (Altındağ et al., 2005; Hekimoğlu et al., 2016). “Honour-related” violence is also reported among women in Muslim-majority countries in the Middle East and in communities in Europe (Douki et al., 2007; Rezaeian, 2010; Canetto, 2015). From Durkheim`s fatalistic model of suicide Attempted Suicide and Suicide ― 121 it could be argued that, Turkish women`s suicidal behaviour is a reaction to the harsh moral demands that are upheld through force (van Bergen et al., 2008). When strict moral rules are experienced by women as external, limiting, demanding and obtrusive, they may experience hopelessness and, ultimately, suicidal thoughts (Durkheim, 1896, 1952). In light of these, the consistency of the “honour-related” violence among this specific group of Turkish women both in Europe and Turkey might be indicative of the continuity of fatalistic suicide. It could also be that the suicidality of Turkish women is not fully accounted for by the preceding fatalistic model. Traditionalism seems to persists especially among the Turkish migrants in Germany, Belgium and the Netherlands, many of whom came from rural areas (e.g. Ersanilli, 2010; Güngör, 2008). Research on Turkish migrants in Europe points to the culture conflict especially when the social role one fulfils is incompatible 5 with the demands and wishes of the majority culture (Schouler-Ocak, 2015; Eylem et al., 2019). It could be that the culture conflict is stressful for Turkish women since there is a great disparity between the demands of the family (i.e. maintaining family “honour”) and those of the host society (i.e. participation of females in social life; Eylem et al., 2019; Schouler-Ocak, 2015). Culture conflict might result in migration-related violence (i.e. social isolation of women by prohibiting friendships and/or their participation in labour force of the host society), disadvantaged socio-economic position and affronts to basic needs (e.g. adequate housing, schooling, access to health care). Thus, the intersection between these social issues, gender and migration related factors might be central to the suicidal behaviours of Turkish women (Sher & Vilens, 2010; McKenzie & Bhui, 2007).

Present study The heightened risk for suicidal behaviours among Turkish women living in Turkey and Europe is a serious public health problem. Even though there is a growing literature on Turkish women`s suicidality, comprehensive systematic reviews are lacking. Thus, in this study, we compare and synthesise the empirical evidence on the characteristics and precipitating factors for suicide and attempted suicide of Turkish women in Europe and 122 ― Chapter 5

Turkey. We investigate whether socio-demographic, migration related (e.g. acculturation), and gender related factors (e.g. “honour-related” violence) are relevant. We also examine the role of mental illness. In this paper, we use the concepts precipitating factors when referring to the perceived causes or important contributors for suicidal behaviour (e.g. different forms of violence), and characteristics when referring to the aspects that increase the likelihood of experiencing suicidal crisis, which may be ‘causal’(e.g. psychiatric or physical illness) and/or not necessarily ‘causal’ (e.g. sociodemographic factors; Heredia Montesinos et al., 2015). Further, the concepts: First generation refers to individuals who were born in Turkey and are settled in the respective host countries, while second generation refers to those who are children of the first generation and were born in the host countries. methods

Identification and selection of the studies

Search Strategy We conducted a comprehensive systematic literature search in the following bibliographic databases: PsycINFO, PubMed, Med Line, Web of Science, Smart Cat, Safety Lit, BASE and Ulakbim. Ulakbim was chosen since it is a comprehensive database including publications in Turkish language from several Turkish databases such as Tübitak, and Sosyal Bilimler. We also checked the reference lists of the included papers in order to identify additional relevant studies up to 27 December 2018. The following terms were used (including synonyms and closely related words) as index terms or free-text words: “suicide”, “suicidal behaviour” or “behaviour”, “suicide attempt”, “self-harm”, “girls”, “women”, “female”, “Turkey” and “Turk”. An additional ‘’Turkish language’’ filter was used to search PsycINFO, PubMed and Smart Cat databases. To retrieve the eligible studies in Turkish, German and Dutch, translations of these key words were used. Attempted Suicide and Suicide ― 123

Inclusion Criteria The searches were limited to the following criteria: 1) Presenting original empirical data on demographic, psychological, psychiatric or social factors for suicide, attempted suicide and self-harm among Turkish girls and women living in Turkey and in Europe 2) Using hospital samples, crisis centre samples (for attempted suicide) and/or forensic reports (for suicide) 3) Being published between 1960 and 2018. Exclusion Criteria The exclusion criteria: 1) Publications which investigated suicidal ideation only, 2) Publications which did not investigate possible gender differences, 3) Publications investigating geriatric populations, 4) Publications which purely reported standardized mortality rates (SMRs), prevalence, or incidence data. Quality Assessment 5 In this review, multiple research designs were employed in line with the multidisciplinary approach recommending mixed methods (e.g. qualitative studies and psychological autopsy studies) to examine the characteristics and precipitants of suicidal behaviours (Hjelmeland, 2013). Since this contributed to the variability of the data, a particular core criteria was used to assess all the studies (see table 1). The core criteria were: 1) Whether female specific information was measured and provided (beyond merely a speculation in the discussion), 2) whether precipitating factors and characteristics of suicide and/or attempted suicide were measured and reported, 3) whether the paper was published in a peer reviewed journal, 4) whether the research questions and the design of the study were clear, 5) in case of a case-control study; whether there was a comparison group from majority ethnic group and/or healthy controls, 6) whether the information on attempted suicide was collected retrospectively only (e.g. through available records), 7) whether the information was gathered from the patients and/or relatives by the researchers directly (through self-report questionnaires and/or interviews), and 8) whether there was any follow-up assessment of the characteristics and/or precipitating factors of attempted suicide. 124 ― Chapter 5

The preceding criteria was decided in consultation with the following: The basic questions and the suggestions for conducting and reporting psychological autopsy studies and other type of studies investigating characteristics and precipitants of suicidal behaviours (i.e. including one or more proxy respondents, including controls, interviewing respondents and controls directly, procedures to describe discrepancies if information was drawn from various sources (Werlang & Botega, 2003; Conner et al., 2012). When the studies did not meet the first four criteria, they were excluded (see figure 1). Two authors (DvB, OE) were involved in the quality assessment. The discrepancies between the authors’ selections were resolved through discussions. If not resolved, the opinion of an independent researcher was sought. Data extraction A customised data extraction form was generated in consultation with the Cochrane Collaborative GRADE approach (Schünemann 2013). This form included, but was not limited to: country/region, population characteristics, sample, place and year of recruitment, methods of data collection, type of suicidal behaviour (suicide or attempted suicide), socioeconomic factors, age, gender, precipitating factors, psychosocial and psychiatric factors. Since there is an evidence for the variability in suicide epidemiology depending on region (Haigh et al., 2016), we have also extracted data on traditionality of the regions of the included studies from Turkey. This was done by one of the first authors (OE) and checked by an independent expert who was a sociologist based in Turkey. The titles of all studies were screened, and the abstracts of the studies were checked regarding our inclusion criteria. When no definitive decision could be made based on the abstract, the original papers were used. Two reviewers (DvB, OE) independently completed extraction forms, and a third reviewer (AHM) established consensus across the two forms. resUlts After examining a total of 9,977 abstracts, we retrieved 322 full text papers for further consideration. We excluded 9,655 of the retrieved papers. The PRISMA flowchart Attempted Suicide and Suicide ― 125 describing the inclusion process, including the reasons for exclusion, is presented in Figure 1. A total of 34 studies were included in the qualitative synthesis and the results are summarised in table 2.

Characteristics of the included studies Of all the included studies, (N=8) 23% were conducted in Europe: (N=3) 9% in the Netherlands, (N=2) 6% in Germany, (N=3) 9% in Switzerland and (N=26) 76% in Turkey. The majority of the studies from Turkey (N=14) 54% were conducted in traditional rural areas, while (N=12) 46% were conducted in less traditional urban areas (see table 2). Regarding the included studies from Turkey, (N=16) 48% investigated attempted suicide, while (N=10) 30% investigated suicide. None of the included studies from Europe investigated suicide. 5 126 ― Chapter 5

Figure 1. PRISMA diagram Attempted Suicide and Suicide ― 127 X X X √ XXXXXX√ XX√ √ √ √ √ X XXX X X XXX √√ √ √ √ √

* * * 5 √ √ √ 1 2 3 4 5 6 7 8 √√ √√ √ √√ √ √ √√ √ √ √ √ √ √ √ √ √ Quality Assessment the of studies or self-report measures) Questions: Whether specific1) information was women on assessed and provided merely (beyond a speculation in the discussion) Whether2) precipitating factors and characteristics attempted suicide were suicide of and/or measured and reported Whether3) the study was published in a peer reviewed journal Whether4) research questions and design the of study were clear the In case5) a case-control of study: Whether there was a comparison from group the majority healthy ethnic controls and/or group 6) Whether the information was collected retrospectively the evaluation the of case only (through reports) Whether7) the information relatives was suicide, the by gathered and/or completed researchers from those attempted and/or who interviews (through and/ Whether8) assessment there was any follow-up to re-assess characteristics precipitating factors and/or attempted of suicide Study Brückner et al. (2008 & 2011) (2009) Bergen Van Burger et al. (2010) Burger et al. (2013) Heredia Montesinos et al. (2018) Yilmaz & Riecher-Rossler (2008, 2012) Aichberger et al. (2015) Table 1. Table 128 ― Chapter 5 X X X X X X X X NA √ √ XX√ XXXX√ √ XXXX√ √ XXXX√√ √ X √ √ √ X √ √ √ √ √ XXXXX XXX X XX X X XXXXX X XXXXX √√ √ √ √ √ * * * √ √ √ √√ √√ √√ √ √√ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √√ √ √√ √ √√ √ √ √ √ √ √ √ √ √ √√ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ Continued.

Akın, and Çil (2007) Tüzün Yektaş et al. (2014) Yektaş Cetin et al. (2001) Devrimci-et al. (2003) Erşan and Kılıç (2013) Konkan et al. (2014) Köse et al. (2012) Senol et al (2005) Simsek et al. (2013) Altindag et al. (2005) Eroglu et al.Eroglu (2013) et al.Guloglu, (2009) Ibiloglu et al. (2016) Ozdel et al. (2009) Saraçoğlu, et al. (2014) et al. (2011) Turhan Yasan, (2008) Table 1. Attempted Suicide and Suicide ― 129 NA NA NA NA NA NA NA NA NA X X X X X X X X X √ √ √ √ √ √ √ √ √ X X X X X X X X X √ √ √ √

* * 5 √ √ √√ √√ √√ √ √√ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ Continued.

: Yes; X: No; *: Social *: X: relationship precipitants No; (i.e. and family: Yes; domestic problems, violence and sexual were others) related not abuse, honour violence and/or Asirdizer et al. (2010) Enginyurt et al. (2014) Goren et al. (2004) Hekimoglu et al (2016) Karbeyaz et al. (2013) Karberyaz et al (2016) Oner et al (2014) Oner et al (2007) et al. (2013) Taktak √ assessment the of precipitants (follow-up and reported; Applicable the characteristicsinvestigated Not and/or NA: suicide to the of attempt is applicable not studies investigating suicide) Table 1. 130 ― Chapter 5

Quality assessment of the included studies Overall, the quality of the studies in the literature was relatively poor. A majority of the retrieved studies (N=124) 38% were excluded based on the following reasons: Not having a clear research question and/or design, not peer reviewed, not reporting gender specific characteristics and/or precipitating factors for suicide and/or attempted suicide (see figure 1). The majority of the included studies (N=21) 62% collected data retrospectively from hospital, public or forensic records. There were three (9%) case-control studies and one (3%) prospective study (see table 1). Furthermore, all but one (Senol et al., 2005) collected information either through direct interviews with the informants (i.e. suicide attempter and/or relative of the attempter and/ or completer) or through evaluation of the available records. Senol and colleagues employed mixed methods (i.e. interviews with the attempters and their hospital records; Senol et al., 2005). However, they did not describe any procedures to deal with the discrepancies in information. All the included studies reported on the characteristics of suicide and/or attempted suicide, while precipitating factors were reported in (N=4) 50% studies from Europe, (N=9) 56% studies from Turkey investigating attempted suicide and (N=6) 60% studies from Turkey investigating suicide (see table 2). “Honour-related” violence was the least investigated precipitating factor. Only (N=2) 25% studies from Europe and (N=2) 20% from Turkey investigating suicide provided evidence for “honour-related” violence. The evidence for this factor was only speculative in (N=9) 56% studies from Turkey investigating attempted suicide (see table 2). Attempted Suicide and Suicide ― 131 Psychiatric disorders (78,6%) disorderAdjustment (49.7%) (22.0%) Depression Psychotic personality or disorders (13.8%) Prior conflict with their spouseownor second 61%, generation: (first family generation: 34.6%) Conflict with theirparents (more common among second generation) Mental illness (13.8%) problems (10.7%) Work-related Difficulties with coping physicalwith illness (1.2%), Domestic violence (0.6%) Financial strains (0.6%)

Findings Characteristics Illnesses Mental · · · · Precipitating Factors · · · · · · ·

5 Methods Surveillance study. Data collection by emergency unit staff, psychiatrists. and Standardized form documentation (socio-demographics, clinical data, and reasons the for attempt). Place and year of Recruitment 10/2008-09/2011 Emergency 40 departments, 3 psychiatric clinics Sample N=159 female Turkish migrants Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Berlin, Germany Berlin, Attempted and completed suicide in females in Turkey and inAttempted Europe and suicide (2000-2018). in completed females in Turkey Authors Aichberger et al. (2015) Table 2: Table 132 ― Chapter 5 Young age (Turkish women aged women (Turkish age Young between years) & 35-39 15-24 Affective (51.4%) disorder Stress-related somatoform disorders were common more compared (45.7%) and Swiss & 25%) men (40%, to Turkish females & 30,6%) (40.6% Personality were disorders vs 10%) (2.9% men commonmore compared to Turkish and Swiss females vs 35.4%) (2.9% females aged 15-25 Turkish age Young the in 310/100.000 vs. (575/100.000 Dutch group) differenceNo in socioeconomic living conditions compared to Dutch

Findings Characteristics Demographic factors · Illnesses Mental · · · Characteristics Demographic factors · · Methods Surveillance study. Data collection hospitalby staff. Standardized anonymized form documentation (clinical and socio- demographic data). Surveillance study. Registration form filled hospital by out staff. Staff askedquestions to patients about demographics and reasons for attempt. Place and year of Recruitment 2003-2004 Hospitals in Basel Kanton that participated in the WHO/ EURO -Multicentre study 2002-2004 Hospitals located in The Hague and its suburbs Sample N=56 Turkish Turkish N=56 immigrants females N=36 Native N=291 Swiss N= 210 females N=1458 suicide N=1458 attempts suicide N=128 cases Attempted suicide Type of Suicidal of Type Behaviour Attempted suicide Country, Region Country, Basel, Switzerland The Hague, The Netherlands Continued.

Authors Brückner et al. (2011) Burger et al. (2010) Table 2: Attempted Suicide and Suicide ― 133 Mental illnesses were less common depression)(except compared to Dutch (7%women vs. 36%) Level high of education among first generation suicide of attempters (0%), second generation of suicide attempters (20%) generation first in (40%) Unemployed in secondand generation (30%)

Findings Characteristics Illnesses Mental · Characteristics Demographic factors · ·

5 Methods Surveillance study Registration form filled hospital by out staff.Staff asked questions to patients about demographics, diagnoses, and reasons for attempt. Focus groups and questionnaire (socio- demographics) Place and year of Recruitment 2008-2010 Hospitals located in The Hague and its suburbs 2011-2015 Recruited through psychiatric clinics, practitioners, psychiatrists, psychotherapists, and women´s shelters Sample N=1534 Dutch N=842 N=209 Surinamese Turkish N=137 N=25 Antillean N=15 attempters, women N=20 from Turkish community Attempted suicide Type of Suicidal of Type Behaviour Attempted suicide Country, Region Country, The Hague, The Netherlands Berlin, Germany Berlin, Continued.

Authors Burger et al. (2015) Heredia Montesinos et al. (2019) Table 2: 134 ― Chapter 5 Affective disorders(73.3%) Neurotic/stress-related somatoform disorder (20%) familyImpact of and community: pressures to collectivism due (family- centeredness leading to neglect self- of and to familycare due and autonomy) culture in Turkish honour The impact German of society: racism, discrimination and lack of acceptance

Findings Mental Illnesses Mental · · Precipitating Factors · · Methods Place and year of Recruitment Sample Type of Suicidal of Type Behaviour Country, Region Country, Continued.

Authors Heredia Montesinos et al. (2019) Table 2: Attempted Suicide and Suicide ― 135 Psychiatric personality or disorder Moroccan 59%, (Dutch 79%, Turkish 60%, Suriname 71%) Co-morbidity (Dutch Suriname 59%, 28%, Moroccan 27%) 33%, Turkish

Findings Characteristics Illnesses Mental · ·

5 Methods Retrospective evaluation of the hospital crisis centre (continues records next page). Place and year of Recruitment 1996-2005 Hospital record case and/or files from VU University Hospital Emergency Unit and a Crisis Centre that mentioned a suicide attempt were included Sample N=115 (female N=115 only sample aged 12-40) Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Amsterdam, The Netherlands Continued.

Authors Bergen Van et al. (2009) Table 2: 136 ― Chapter 5 In at least at In half the of cases, specific stressful life events related to family honour and personal autonomy reported were restriction Moroccan 31%, Physical abuse (Turkish 37%, Suriname 42%, Dutch 21%) Sexual Moroccan 19%, abuse (Turkish 17%, Suriname Dutch 21%, 38%) Overregulation: cases of 55% for and SurinameMoroccan, vs. Turkish Dutch: 28% Demand maintaining for sexual & Moroccan) Turkish abstinence (30% Fear being and of 9% an outcast (Turkish Moroccan) 13% Being to maintain forced unwanted marriage (Dutch 0%, Suriname 0%, 6%. Moroccan 10%) Turkish Rejection partner of (Dutch 3%, Suriname 9%, Moroccan 10%) 25%, Turkish

Findings Precipitating Factors · · · · · · · · Methods Medical case files files case Medical filled hospital by out staffor crisis centre staff. Clinical interviews were held with and women reasons attempt for were asked for. Sometimes medical expert reports were added to the case file. Place and year of Recruitment Sample Type of Suicidal of Type Behaviour Country, Region Country, Continued.

Authors Table 2: Attempted Suicide and Suicide ― 137 Young age Turkish females aged 15-24 Turkish age Young attempts of among Turkish (38.6% immigrants) Domestic violence within family and 14.7% generation, first (24.1% partnership second generation) partnership within violence Domestic females) Turkish (18%

Findings Characteristics · Precipitating Factors · ·

5 Methods Surveillance study. Data collection medicalthrough files and evaluation form filled by out physician attending (socio-demographics, main reasons the for attempts, diagnoses). Place and year of Recruitment 1991-1997 Emergency Unit Universityof Hospital of Basel-City Sample N=140 Turkish N=70 N=70 Swiss Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Basel, Switzerland Continued.

Authors Yilmaz and Riecher- Rossler (2008) Table 2: 138 ― Chapter 5 Suicide attempt rates among Turkish Suicide attempt rates among Turkish immigrants times higher 2.7 than among Swiss males, 36% females 64% Turkish: Swiss: 28% males, females 72% femalesHighest rates among Turkish and 35-39 aged 15-24 AffectiveTurkish females disorders: 51.4 vs. Swiss 30.6% females disorder:Adjustment Turkish 45.7% females vs. 14.6% Swiss females

Findings Characteristics Demographic factors · · · · illness Mental · · Methods Surveillance study. Data collection hospitalthrough Standardized staff. anonymized form documentation (clinical and socio- demographic data). Place and year of Recruitment 2003-2004 Emergency Unit Universityof Hospital of Basel-City WHO–EURO Multicentre Study on Suicidal Behaviour Sample N=271 Turkish N=46 N=225 Swiss Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Basel, Switzerland Continued.

Authors Yilmaz and Riecher- Rossler (2012) Table 2: Attempted Suicide and Suicide ― 139 economically / dependent their on husbands* Female attempts of gender females) by (69% 35%: (vs group 57% age age in15-25 Young 25 and according older to the Turkish Statistics) of Ministry Domestic violence (24%) Being unemployed

Findings Characteristics · · Precipitating Factors · ·

5 Methods Clinical interviews with the patients admitted to the hospital Place and year of Recruitment 2005-2006 Diçle University Hospital Sample N=80 females N=55 N=25 males Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Diyarbakir, South-East Turkey, traditional Continued.

Authors Akın, & Tüzün (2007) Çil Table 2: 140 ― Chapter 5 Depression was common more in female psychiatric had not outpatients (who thanattempted) males in the other two groups* Negativity in the familial aspect the of problems withself-image the family (i.e. was often more reported females. by self-image inNegative the family was oftenmore reported females by and seen as an important factor separating suicide attempters from non-attempters)*

Findings Characteristics · Precipitating Factors · Methods Case-control study Demographic information and self-reported data from patients their on were health mental obtained through validated scales. Place and year of Recruitment Data collection period was not specified Child, Adolescent and Adult Psychiatry Departments of Hacettepe University Center Medical Sample N= 33, N=23 N=23 N= 33, females, males N=10 (youth) Attempters, did obtain not health mental treatment N=50 (N=33 females, N=17 controlsmales) receivedwho health mental treatment controls N=50 females, (N=26 N=24 males) Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Ankara, Central Central Ankara, Anatolia, less traditional Continued.

Authors Cetin et al. (2001) Table 2: Attempted Suicide and Suicide ― 141 Female gender (69.74% of attempts of by Female gender (69.74% females vs 30.26% males) females (1999: 64.93 (15-24) % vs age Young 35.07%; 2000: females 69.32% vs 30.60%; 2001: females vs 29.47% 70.53% males) females ( 68% vs males) Unemployment 32% participatingNot in (57% labour force females were housewives mostly and college students) Female gender (female to male 3:1)* ratio: had a female 15-24 group (age age Young to male 3.6:1)* ratio: Being married females (42% vs. % males) 31 females males) vs. (64% 31% Unemployment females (30% males) vs. problem Family 21% femalesDomestic violence (9% vs. males) 2%

Findings Characteristics · · · · Characteristics · · · · Precipitating Factors · ·

5 Methods Retrospective evaluation of the records. hospital Hospital staff administered monitoring forms of patients treated for attempted suicide Clinical interviews with the patients admitted to the hospital Place and year of Recruitment 1998-2001 5 hospitals & 22 primary care units WHO–EURO Multicentre Study on Suicidal Behaviour 2006-2008 17 emergency rooms state of hospitals Sample N=737 females N=514 N=223 males N=893 N=660 females N=230 males Attempted suicide Type of Suicidal of Type Behaviour Attempted suicide Country, Region Country, Mamak- Ankara, Ankara, Mamak- Anatolia, Central less traditional Erzurum & Erzincan province, North Eastern Anatolia, traditional Continued.

Authors Devrimci- et al. (2003) Eroglu et Eroglu al. (2013) Table 2: 142 ― Chapter 5 Gender females (70% vs 30 % males) Being %) married (0.55 Being single (0.43 % males only) femalesHigher lethal intent (61% vs. 39% males) Female gender females (72% vs 28% males) 63%) 15-24: group (age age Young Economic females dependence (34% were housewives) History mental of females, illness 24 (16% % males) Mental illness females males) vs. (52% 31% Marital problems (77% females vs. 20% males) Relationship problems with family and a spouse/partner (77%) Domestic violence (18%)

Findings Characteristics · · · · Characteristics · · · Illnesses Mental · · Precipitating Factors · · · Methods Retrospective evaluation of the records hospital Standardized anonymized forms documentation were filled by out staff attending Retrospective evaluation of the records hospital Standardized anonymized forms documentation were filled by out staff attending Place and year of Recruitment 2011-2012 Emergency Department of Sivas Numune Hospital 2003-2007 Dicle University Hospital Sample N=291 (210 (210 N=291 females, 81 males) N=1281 femalesN=901 Attempted suicide Type of Suicidal of Type Behaviour Attempted suicide Country, Region Country, Sivas, Central Anatolia, traditional Diyarbakir South East Turkey, traditional Continued.

Authors Erşan & Kılıç (2013) Guloglu, et Guloglu, al. (2009) Table 2: Attempted Suicide and Suicide ― 143 Relationship problems with their spouse females) familyor (42.5% emotions “focus on of & ventingUse of considered to be a emotion strategy”, non-functional problem-solving strategy (Means: 13.27 among females vs. 11.58 males)* among

Findings Precipitating Factors · Precipitating Factors* ·

5 Methods Clinical interviews with the patients admitted to the hospital Case-control study Self-reported data from patients their on were health mental obtained through validated scales Place and year of Recruitment Data collection period was not specified Department of Psychiatry at Dicle University Faculty of Medicine 2010-2011 Bakırkoy Prof. MazharDr. Osman Mental Health and Neurological Diseases Hospital Sample N=106 females N=57 N=49 males Ages: 18-55 N= 102 femalesN=54 N=48 males Attempted suicide Type of Suicidal of Type Behaviour Attempted suicide Country, Region Country, Diyarbakir South East Turkey, traditional İstanbul, Western less Turkey, traditional Continued.

Authors Ibiloglu et al. (2016) Konkan et al. (2014) Table 2: 144 ― Chapter 5 Female gender (82%) age-females years aged 15-24 (71%) Young participatingNot in (i.e. labour force being a housewife, 35 %) problems (45%) Family Domestic violence (17%) Loneliness (5.3%) Illiteracy females (37% vs. % males) 19 Marital females status (36.1% vs 38.9% males) Low religious females orientation (86.1% males) vs 81%

Findings Characteristics · · · Precipitating Factors · · · Characteristics · · Precipitating Factors · Methods Retrospective evaluation of the records hospital Standardized anonymized forms documentation were filled by out staff attending Clinical interviews with the patients admitted to the hospital Place and year of Recruitment 2009 Education Van & Research Hospital 2006-2007 Emergency Clinic of Pamukkale University Hospital Sample N=112 femalesN= 92 N=20 males N=144 suicide N=144 females N=108 males N=36 Attempted suicide Type of Suicidal of Type Behaviour Attempted suicide Country, Region Country, Van, Eastern Van, Turkey, traditional Pamukkale, Turkey, Western less traditional Continued.

Authors Köse et al. (2012) Ozdel et al. (2009) Table 2: Attempted Suicide and Suicide ― 145 Female gender (69%) Depression females (72% vs. 27% males) Anxiety females (65% vs. 35% males) females disorderAdjustment (61% vs. 38% males) Bipolar females disorder males) vs. (56% 45% Borderline Personality Disorder (86% females males) vs. 14% Other psychological disorders (67% females vs. 33% males)

Findings Characteristics · Illnesses Mental · · · · · ·

5 Methods Clinical interviews with the patients admitted to the hospital Place and year of Recruitment 2009-2011 Çukurova University Faculty of Medicine Sample N=122 femalesN=84 N=38 males Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Cukurova, Cukurova, South Turkey, traditional Continued.

Authors Saraçoğlu et al. (2014) Table 2: 146 ― Chapter 5 Female gender (63%) age-females years aged 15-24 of Young females (66% age vs. males) 34% Being married (48.3%) Being males single (69.7% only) Low education (49%) participatingNot in (i.e. labour force being a housewife, 62%) Domestic violence (50%) Other precipitants related with gettingrelationship problems (e.g. (63% beingdivorced, not to conceive) able females vs. 7% males) conflictsFamily females(74% vs. males) 26% School females failure (65% vs. 35% males) failure (67% femalesJob vs. 33% males)

Findings Characteristics · · · · · · Precipitating Factors · · · · · Methods Mixed methods (clinical interviews with the patients & evaluation of the records) hospital Place and year of Recruitment 2001-2002 Erciyes University, Faculty of Medicine, Acute and Emergency Department Sample N=333 (209 (209 N=333 Females, 124 Males) Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Erciyes, Central Anatolia, traditional Continued.

Authors Senol et al. (2005) Table 2: Attempted Suicide and Suicide ― 147 Female gender (77%) , female to male ratio (3.47:1) conflict femalesFamily (30% males) 21% vs. femalesDomestic violence males) (9% vs. 4% Relationship problems within the family females(29.7% males) vs. 21% Female and 15-24 age young gender (78,8%) Mental illness females (9% vs. males) 13% Relationship problems with opposite females vs.gender 17% (14% males) femalesDomestic % vs. 12 violence (14% males) femalesEconomic problems vs. (2% males) 8% Relationship problems within the family females(42% vs. 33% males) femalesRape males) vs. 0% (0,5%

Findings Characteristics · Precipitating Factors · · · Characteristics · Illnesses Mental · Precipitating Factors · · · · ·

5 Methods Descriptive surveillance study Hospital staff administered monitoring forms of patients treated for attempted suicide. Retrospective evaluation of the records. hospital Hospital staff administered monitoring forms of patients treated for attempted suicide. Place and year of Recruitment 2010 Emergency Service suicide registryattempt record forms public from 12 and private hospitals 01/2007-12/2009 Emergency services 8 state of hospitals Sample N=693 % female77.6 33.4 % male N=1613 N=1270 females males N=343 Attempted suicide Type of Suicidal of Type Behaviour Attempted suicide Country, Region Country, Sanliurfa, South-East Turkey, traditional Hatay, South Hatay, East Turkey, traditional region Continued.

Authors Simsek et al. (2013) Turhan et Turhan al. (2011) Table 2: 148 ― Chapter 5 Female gender (69%) were age: 73% 15-24 group age of Young females Low education females (76% had primary education only) (87% female) Unemployment Marital status females (27% vs 33.3% males) Depression both genders, for was rate 51% co morbidity psychiatric of disorders females males) vs.(24% 6% Religion: both genders were less devoted than the rest their of household. Stressful forced events within family (e.g. into undesirable marriage, discouraged from females seeking (62% employment) males) vs. 18%

Findings Characteristics · · · · · Illnesses Mental · · Precipitating Factors · · Methods Prospective study Clinical interviews with the patients admitted to the hospital. Place and year of Recruitment 02/2005-07/2005 University Hospital Dicle,of Diyarbakir 2006: follow- a through up telephone survey Sample N=96 subjectsN=96 who attempted suicide by poisoning first the for time and 15 (age above) 76 cases (51 females, 25 weremales) reached the at after follow-up one year. Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, Diyabakir, South- East Turkey, traditional Continued.

Authors et al. Yasan (2008) Table 2: Attempted Suicide and Suicide ― 149 Unfavourable attitude their of family Unfavourable females vs.(64% males) 36% Lack familial of social or support females vs.8% (31% males) violenceFamily after the attempt (65% females vs. males) 24% Persistence trigger of females factors (33% males) vs. 12% Lack access of females to therapy (63% vs. 40% males)

Findings Follow-up of the precipitating factors precipitating the of Follow-up · · · · · ·

5 Methods Place and year of Recruitment Sample Type of Suicidal of Type Behaviour Country, Region Country, Continued.

Authors et al. Yasan (2008) Table 2: 150 ― Chapter 5 Relationship problems with parents (n=3.75%) 25%) Loneliness (n=1, Relationship problems with parents (n=5, 31.25%)

Findings Precipitating Factors Precipitating factors females for with no serious intention to die (n=4): · · Precipitating factors females for with indefinite intention to die (n=16) · Methods Cross-sectional study Self-reported data from patients their on were health mental obtained through validated scales Place and year of Recruitment Data collection period was not specified UniversityEge Faculty of Medicine Department Childof and Adolescents Health Mental Sample N=79 females N=79 aged 15-17 years Attempted suicide Type of Suicidal of Type Behaviour Country, Region Country, İzmir, Western Western İzmir, less Turkey, traditional Continued.

Authors et Yektaş al. (2014) Table 2: Attempted Suicide and Suicide ― 151 Problems at school (i.e. academicProblems school at (i.e. failure) (n=3, 18.75%) Relationship problems with friends (n=2, 12.5%) Problems school at & relationship 12.5%) problems with friends (n=2, Relationship problems with family & 25%) friends (n=4, Relationship problems with parents (n=8, 50%) Relationship problems with friends (n=2, 12.5%) Relationship problems with family & 25%) friends (n=4, Conflicts police, (e.g. authoritywith the 6.25%) school staff)(n=1, Conflicts with friends,family & 6.25%) authority (n=1,

Findings · · · · Precipitating factors females for with definite/serious intention to die (n=19) · · · · ·

5 Methods Place and year of Recruitment Sample Type of Suicidal of Type Behaviour Country, Region Country, Continued.

Authors et Yektaş al. (2014) Table 2: 152 ― Chapter 5 Female gender (female to male ration: 1.7:1)* third (two age was between 15-24 Young years old). participating not Illiteracy, in labour having not paid work)* (i.e. force girls are related not Honour violence (e.g. sent to school their activities are limited to the home).

Findings Characteristics · · · Precipitating Factors* · Methods Controlled psychological autopsy study Place and year of Recruitment Batman Public Prosecutor 2000 Sample N=31 (total (total N=31 cases suicide of in Batman in 2000 female n=26 cases, age an (with 15-24 informant. 5 cases,In no informants was willing to participate). 25 control Plus cases Suicide Type of Suicidal of Type Behaviour Country, Region Country, Batman, Southeastern Turkey, traditional Continued.

Authors Altindag et al. (2005) Table 2: Attempted Suicide and Suicide ― 153 Illness (31.34% femalesIllness vs 28.42% (31.34% males) femalesEconomic problems (5.17% vs 19,0% males) Relationship problems with the family incompatibility(i.e. with the family) (females: males 31.29%, 16.82%) Relationship problems with a partner/ failedspouse romantic (i.e. relationship) females vs(3.82% 2.84 % males) Educational Failure(5.46 % females, 5.46 % males).

Findings Characteristics · · Precipitating Factors · · ·

5 Methods Suicide statistics were retrieved from the Turkish Institutes Statistical (TURKSTAT) Place and year of Recruitment The entire entire The country (Turkey) was covered. Years:1995-2005 Sample N=22.350 (n= 13692 males and n=8658 females) Suicide Type of Suicidal of Type Behaviour Country, Region Country, Inhabitants of the Republic of Turkey Continued.

Authors et Asirdizer al. (2010) Table 2: 154 ― Chapter 5 Being females) illiterate (7.5% finishedLow education (not a primary % females males) % of vs (13.9 7.8 school) males). vs 18,8% Illness (22.5% females problems (14.3% Family vs 9.1 % males) unrequitedRelationship problems (i.e. females only) (5.5% love) age-females below 20 the of age Young (56.3%) Being married females) (37% (interpretation: hymen’ ‘Deflorated relatedhonour issue in traditional family structures) females) (13%

Findings Characteristics · · · Precipitating Factors · · Characteristics · · Precipitating Factors · Methods Suicide statistics were retrieved from the Turkish Institutes Statistical (TURKSTAT) Post-mortem investigation and reports autopsy Place and year of Recruitment The entire entire The country (Turkey) was covered. 2007-2012 Years Diyarbakir Province 1996-2001Years: Sample N=17342 males n=12107 and 5235 females N=302 females n=174 n=128 Suicide Type of Suicidal of Type Behaviour Suicide Country, Region Country, Inhabitants the of Republic Turkey Diyabakir, South- East Turkey, traditional Continued.

Authors Enginyurt et al. (2014) Goren et al. (2004) Table 2: Attempted Suicide and Suicide ― 155 Young age-females aged betweenYoung 16-20 (45.45 %) participatingNot being in (i.e. labour force % ) (87.9 household) a housewife/doing Being under treatment depression for 7.6% Being married (33.3%) Being married as a fellow wife %) (4.5 Being married bride through exchange the of nine cases (one %) under the (15.2 was married 15 of age bride through exchange) Physical signs domestic of violence (18.2%)

Findings Characteristics · · Illnesses Mental · Precipitating Factors · · · ·

5 Methods Post-mortem investigation and reports autopsy Place and year of Recruitment (Turkey) was (Turkey) covered 2005-2011 Years: Sample N=66 (females)N=66 City Van Suicide Type of Suicidal of Type Behaviour Country, Region Country, Van, Eastern Van, Turkey, traditional Continued.

Authors Hekimoglu et al. (2016) Table 2: 156 ― Chapter 5 Possible diagnosis a mental of illness females) (64% Being (73% unemployed females vs 27% males) A formal diagnosis a mental of illness females(47% vsmales) 22% Loneliness females (35% vs males) 23% Relationship problems with a partner females(27% vs 17% males) Loss females a relative of % vs 11 (21% males)

Findings Characteristics · Characteristics · Illnesses Mental · Precipitating Factors · · · Methods Post-mortem investigation and reports autopsy Post-mortem investigation and reports autopsy Place and year of Recruitment The city of 1997- Eskisehir, 2011 Eskisehir City Eskisehir 2004-2015 Sample N=553, n=395 n=395 N=553, male and 158 were female N=75 females, n=34 n=41 males suicide(75/428 cases were students) Suicide Type of Suicidal of Type Behaviour Suicide Country, Region Country, Eskisehir, Eskisehir, Western Anatolia, less traditional Eskisehir, Eskisehir, Western Anatolia, less traditional Continued.

Authors Karbeyaz et al. (2013) Karberyaz Karberyaz et al. (2016) Table 2: Attempted Suicide and Suicide ― 157 Higher proportion females of aged 15-24* The leading reasons suicide in for females was relationship problems with family and/ spouse* Other reasons (business failure, illness, educational were failure) less common among females compared to males.(not downbroken age)* by

Findings Characteristics · Precipitating Factors · ·

5 Methods Post-mortem investigation of annual records of National Turkish Institute Suicide of Statistics Place and year of Recruitment The entire entire The country was covered (Turkey) Sample N=44586 N= 16249 females (36%) N= 28347 ) males (64% Suicide Type of Suicidal of Type Behaviour Country, Region Country, Inhabitants the of Republic Turkey Continued.

Authors Oner et al. (2014) Table 2: 158 ― Chapter 5 Young age-females aged under 15 Young and 95% CI.55;90) P<0.001, (z=8.06, CI P<0.001 between (z=36.56 15-24 age 2.64: 2.94). % femalesIllness in all (6.90 % to 11.52 groups % males) age vs 8.02 % to 11,24 unsatisfactoryRelationship problems (i.e. relationships) % to 14.26% (5.47 females % to 11.48% males). Economicvs 6.15 to 15.99% femalesproblems (3.39% vs 5.22% to 14.26% males to 13.25 % Educational (3.31% Failure females to males 11.46% vs 6.59

Findings Characteristics · · Precipitating Factors · · Methods Post-mortem investigation of annual records of National Turkish Institute Suicide of Statistics Place and year of Recruitment The entire entire The country was covered (Turkey) Sample N=17.327, males: n= 10585, females = 6742 Suicide Type of Suicidal of Type Behaviour Country, Region Country, Inhabitants the of Republic Turkey Continued.

Authors Oner et al. (2007) Table 2: Attempted Suicide and Suicide ― 159 Young age (mean 29 (mean years, age standardYoung deviation males 13.8/ mean 35.5, age standard deviation 17.0)* Being married (44 % females vs 49% males) femalesBeing (53.6% vs 46.4 unemployed % males). A formal diagnosis a mental of illness females males) (32.8% vs 67.1% treatmentUnder a mental of illness (50% males females) vs 50%

Findings Characteristics · · · Psychiatric physical of illness · ·

5 Methods Post-mortem investigation and reports autopsy Place and year of Recruitment Istanbul Forensic Medicine Institute April- August 2002 Sample N=124, n=83 male and n=41 female Suicide Type of Suicidal of Type Behaviour Country, Region Country, İstanbul, Western less Turkey, traditional Continued.

Authors et Taktak al. (2013) Table 2: *Percentages were not investigated and/or provided*Percentages in were the paper investigated not and/or 160 ― Chapter 5

characteristics of sUicide and attempted sUicide among tUrkish Women in tUrkey and in eUrope

Socio-Demographic Characteristics

Gender Turkish women were overrepresented in studies of attempted suicide in both Europe and Turkey. For instance, among the first generation Turkish immigrants in Basel, the sex ratio of female to male was 1.3:1, and among the second generation it was 3.1:1 (Brückner et al., 2008; 2011). Two studies (Köse et al., 2012; Turhan et al., 2011) with the steepest gender ratio for attempted suicide (approximately 80% females; Köse et al., 2012; Turhan et al., 2011) were conducted in more traditional areas in Turkey. Contrary to these findings, majority of the studies of suicide (Asırdızer et al., 2010; Enginyurt et al., 2014; Karbeyaz et al., 2013; Karberyaz et al.2016; Öner et al. 2014; Oner et al., 2007; Taktak et al., 2013) indicated that men died more often by suicide than women in Turkey, with one exception (Altındağ et al., 2005). Altindag and colleagues reported on female to male ratio of suicide around 1.7:1, indicating higher suicide rates among women compared to men (Altındağ et al., 2005) in a traditional area in Turkey. Age The most frequent age group that presented with a suicide attempt at the hospital was between 15 and 24 both in Turkey and Europe. One study indicated that Turkish womens below the age of 25 accounted for the 38.6% of the attempted suicide cases in Switzerland (Yılmaz and Riecher-Rössler 2008). Similar findings were reported in Turkey with respect to the age group of women attempting suicide (Akın, Tüzün & Çil, 2007; Eroglu et al., 2013; Erşan & Kılıç, 2013; Köse et al., 2012; Turhan et al., 2011; Yaşan, 2008) and completing suicide (Altındağ et al., 2005; Gören et al., 2004; Hekimoğlu et al., 2016; Karberyaz et al., 2016; Öner et al., 2014; Öner et al., 2007; Taktak et al., 2013) when compared to men in traditional areas. Adolescents were also overrepresented in studies investigating suicide in Turkey. For instance, the data derived from the annual Attempted Suicide and Suicide ― 161 reports of suicide statistics of Turkey published between 1998 and 2000 indicated a high number of suicides among girls below the age of 15 (Öner et al., 2007). Social/education status Overall, contradictory findings were reported in the literature with regards to the social and education status of women who attempted and/or completed suicide in Turkey and Europe. One study of attempted suicide from Germany reported that 20 % women of second generation aged 18 and 34 had higher level of education, while none of the women of first generation aged 35 and 49 and/or 50 and above had higher education (Heredia Montesinos et al., 2018). Regarding Turkey, a considerable number of the studies of attempted suicide reported low levels of education among women regardless of the age (Akın, Tüzün & Çil, 2007; Erşan & Kılıç, 2013; Köse et al., 2012; Saraçoğlu, et al., 2014; Şenol et al., 2005; Şimşek 5 et al., 2013; Yaşan, 2008). Furthermore, three studies of suicide (Altindag et al., 2005; Enginyurt et al., 2014; Taktak et al., 2013) and one study of attempted suicide (Köse et al., 2012) presented findings of illiteracy stratified by gender. These studies found that illiteracy was two to three times higher among women than men, which reflects the gender gap in illiteracy in the general Turkish population. Three studies (Altindag et al., 2005; Taktak et al., 2013; Köse et al., 2012) reported high percentages of illiterate women. For instance, study 28 reported 42% of female victims to be illiterate (women in the general population: 7.5%). However, these high percentages of female illiteracy seem influenced by the area/site of the studies (traditional, rural areas). Marital status Turkish women who attempted and/or completed suicide were more often married than ethnic-majority women in Europe and Turkish men in Turkey. For instance, two studies of attempted suicide (van Bergen, 2009; Yılmaz & Riecher-Rossler , 2008) from Europe indicated that Turkish women were more often married (or partnered) compared to women from ethnic majority groups (62% married Turkish females, vs. 41% Dutch females; van Bergen, 2009; 72 % married Turkish persons, vs. 68% Swiss persons; Yılmaz & Riecher-Rossler , 2008). 162 ― Chapter 5

Regarding Turkey, similar percentages were reported among Turkish women in Turkey when compared with men (Eroğlu et al., 2013; Güloğlu, et al., 2009; Şenol et al., 2005; Turhan et al., 2011). Additionally, two studies (Gören et al., 2004; Hekimoğlu et al., 2016) from traditional areas of Turkey reported that a third to almost half of the female suicide cases had been married (37% ; 33.3), with one exception from a less traditional area (44% females vs. 49% males; Taktak et al., 2013). Employment/ socio-economic status Only one study of attempted suicide (Yılmaz & Riecher-Rössler.,2008) provided information about employment of Turkish women in Europe and found that not participating in labour force was three times higher among first generation Turkish women compared to the women from ethnic majority group in Switzerland (30% vs 10%; Yılmaz & Riecher-Rössler.,2008). Regarding Turkey, similar percentages were reported when women were compared with men in a study from a traditional area (64% women vs 31% men; Eroğlu et al., 2013). A considerable number of the studies investigating attempted suicide (Akın, Tüzün & Çil, 2007; Erşan & Kılıç, 2013; Köse et al., 2012; Senol et al., 2005; Yaşan, 2008) and suicide (Altındağ et al., 2005; Hekimoğlu et al., 2016; Karbeyaz et al., 2013) provided further evidence indicating that not being in labour force was common among female suicide attempters and completers compared to amle attempters. All but one of the studies ( Taktak et al., 2013) were from traditional areas in Turkey. Low socio-economic status was mentioned in four studies (Asırdızer et al., 2010; Hekimoglu et al., 2016; Oner et al., 2014; Oner et al., 2007) of suicide in Turkey. Only one study from a traditional area (Hekimoğlu et al., 2016) reported that 86.4% women aged 16 and 20 were listed as “very poor”, while two population-based studies (Asırdızer et al., 2010; Öner et al., 2014), indicated that economic problems were less often reported among women compared to men who died because of suicide (e.g. 57% vs. 19% in Asırdızer et al., 2010). Attempted Suicide and Suicide ― 163

Religiosity Religiosity was not often investigated as a characteristic of suicidal behaviours in the included studies in this current review. One study of attempted suicide from Germany (Heredia Montesinos et al., 2018) and two studies of attempted suicide from Turkey (Özdel et al., 2009; Yaşan, 2008) investigated religion systematically among Turkish women by means of interviews. Religiosity was not assessed in studies of suicide. Heredia Montesinos and colleagues (2018) found that all Turkish immigrant females identified themselves as Muslim (Heredia Montesinos et al., 2018). However, two studies of attempted suicide from Turkey pointed to the low religious devotion as a characteristic of Turkish women in traditional areas in Turkey (Özdel et al., 2009; Yaşan, 2008). For instance, Özdel and colleagues found that low religiosity was more common among women (94%) than among men (81%). 5

Mental illnesses Depression was reported frequently among Turkish female suicide attempters in Europe and Turkey. Aicherber and colleagues for instance, found that depression was almost two times more common among Turkish female suicide attempters (51%) compared to female suicide attempters from ethnic majority background (33,9%) in Germany (Aicherber et al., 2015). With regards to Turkey, two studies of attempted suicice, with high percentages of depression, were conducted in traditional areas (Yasan et al., 2008; Saraçoğlu, et al., 2014). Saraçoğlu and colleagues indicated that depression was three times more likely to be reported among women (72%) than among men (27%; Saraçoğlu, et al., 2014). Adjustment and stress-related disorders were also assessed among Turkish women who attempted suicide in Germany (49.7%; Aicherber et al., 2015) and Switzerland (45.7%; Brückner et al., 2008; 2011). Additionally, one study of attempted suicide provided evidence on large gender differences in adjustment disorders (61% vs 38%) among women compared to men in Turkey (Saraçoğlu, et al., 2014). Furthermore, although three studies of suicide (Karbeyaz et al., 2013; Karberyaz et al., 2016; Taktak et al., 2013) reported that 164 ― Chapter 5 mental illnesses were more common among women compared to men, the specific type of mental illness was not specified. precipitating factors related With sUicide and attempted sUicide among tUrkish Women in tUrkey and in eUrope

Relationship problems with spouses or family Relationship problems were more often reported as precipitating factors for attempted suicide and suicide among Turkish women than ethnic-majority women in Europe and Turkish men in Turkey. Four of eight studies of attempted suicide from Europe (Aichberger et al., 2015; Heredia Montesinos et al., 2018; Yılmaz & Riecher-Rössler.,2008; 2012) highlighted relationship problems with spouses or family members as precipitating factors. For instance, one study provided evidence for a prior conflict with a spouse as a precipitating factor for attempted suicide among 61% Turkish women in Germany (Heredia Montesinos et al., 2018). Regarding Turkey, relationship problems were reported among a considerable number of the studies of attempted suicide (Cetin et al., 2001; Eroğlu et al., 2013; Köse et al., 2012; Şenol et al., 2005; Şimşek et al., 2013; Turhan et al., 2011; Yektaş et al., 2014), with percentages varying from 25% (Eroğlu et al., 2013) to 45% among women (Köse et al., 2012). Those problems were mostly mentioned in the family context related to the social precipitants such as traditional gender roles and expectations (e.g. the expectation from the females to put the family first). For instance, one study reported that female suicide attempters perceived their role in the family as negative and felt this way significantly more often than young female controls who had not attempted suicide (Cetin et al., 2001). Similarly in studies of suicide, “incompatibility with the family” was reported among 31% women compared to 17% men (Asırdızer et al., 2010). Attempted Suicide and Suicide ― 165

Domestic violence and sexual abuse Domestic violence was often reported among the studies of attempted suicide both in Turkey and Europe. For instance, one study of attempted suicide identified domestic violence as the main precipitating factor for Turkish women in Switzerland (21.4% first generation, 14.7% second generation), whereas none of the female comparisons from the majority ethnic group reported it (Yılmaz & Riecher-Rössler, 2008). Additionally, domestic violence was listed as an important social precipitant for attempted suicide in traditional areas in Turkey (Akın, Tüzün & Çil, 2007; Eroğlu et al., 2013; Erşan & Kılıç, 2013; Şenol et al., 2005; Şimsek et al., 2013; Turhan et al., 2011; Yaşan, 2008). Frequencies varied from 9% (Eroglu et al., 2013; Yasan, 2008) to 50% (Şenol et al., 2005). In terms of the gender differences, domestic violence was 2 times (9% females vs 4% males; Simsek et al., 2013) and 4 times (9% females vs 2% males; Eroğlu et al., 2013) more common among 5 women compared to men in traditional areas in Turkey. Only one study indicated that the gender difference for domestic violence was relatively small (males 12% and females 14%; Turhan et al., 2011). Furthermore, domestic violence was reported as a possibility among 18.2% female suicide cases who demonstrated physical signs of trauma related violence (Hekimoğlu et al., 2016). Sexual abuse was rarely mentioned as a precipitating factor. For instance, one study of attempted suicide from the Netherlands (van Bergen et al., 2009) indicated that sexual abuse was less often reported as a precipitating factor among Turkish women (19%) compared to Dutch women (38%). Furthermore, only one study from Turkey assessed rape as a precipitating factor for attempted suicide and provided very small percentages (0.5% women vs. 0% men ; Turhan et al., 2011). “Honour- related” violence and personal autonomy restrictions In Europe, one study (van Bergen, 2009) indicated that more than half (55%) of Turkish women reported “honour-related” conflict (e.g. being accused of not maintaining sexual abstinence until marriage) as a precipitating factor for attempted suicide. Furthermore, another study from Germany reported that conflicts between Turkish women and their parents and/or spouses were linked to family honour (Heredia Montesinos et al., 166 ― Chapter 5

2018). Moreover, honour-related issues were mentioned in nine out of the sixteen studies investigating attempted suicide in Turkey (Akın, Tüzün & Çil, 2007; Devrimci et al., 2003; Eroglu et al. 2013; Erşan & Kılıç, 2013; Köse et al., 2012; Şenol et al., 2005; Şimsek et al., 2013; Güloğlu, et al., 2009; Altındağ et al., 2005) but was not assessed. All but one (Devrimci et al., 2003) were conducted in traditional regions of Turkey. Additionally, not maintaining sexual abstinence until marriage was reported as a reason for conflict with the family members amongst 13% female suicide cases (Gören et al., 2004). On a relevant topic, personal autonomy restrictions were reported extensively by a qualitative study of Heredia Montesinos and colleagues in Germany (Heredia Montesios et al., 2018). It was found that family and community pressures (i.e. expectation from a women to put family first in personal life choices) in the context of the traditional Turkish family system leads to lack of self-acceptance of women, social isolation and suicide crises (Heredia Montesinos et al., 2018). Furthermore, two studies from Turkey (Yaşan, 2008; Hekimoğlu et al., 2016) provided evidence for autonomy restrictions as potential triggers for conflicts resulting with attempted and completed suicide in traditional areas. Specifically, one study reported that 15.2% women aged 16 and 20, and just over 10% girls under the age of 16 were married through bride exchange (Hekimoğlu et al., 2016). Migration related factors Culture conflict was reported only in a qualitative study of Heredia Montesinos and colleagues (Heredia Montesinos et al., 2018). Turkish women in Berlin often felt they could not meet the socio-cultural expectations of German society, such as being strong, independent and assertive, because they felt that Turkish culture demanded the opposite: that they be submissive and family oriented. Furthermore, Turkish women reported that they perceived discrimination against them and a lack of acceptance from German society as leading to feelings of social exclusion and isolation. Furthermore, women who were recent immigrants mentioned a lack of social support (being away from their natal family) and language difficulties as important acculturation-related stressors (Heredia Montesinos et al., 2018). Attempted Suicide and Suicide ― 167

discUssion The results of this systematic review highlight rather similar characteristics and precipitating factors in all the included studies, regardless of the country/continent (Turkey versus Europe) or the type of suicidal behaviour (attempted suicide or suicide). Turkish women who demonstrated suicidal behaviours were relatively young and married at a relatively young age. They were often not part of the labour force, tended to have low socio-economic status, presented symptoms of mental illnesses, such as depression, and were often victims of violence and family conflicts. The cross-national consistency of the characteristics and precipitants of suicidal behaviours among young Turkish women in Europe and Turkey provides further support for the continuous influence of socio- demographic affronts and violence- and interpersonal-conflict-related factors that exist in pre- and post-migration contexts. 5 One important finding was the possible link between Turkish women`s suicidal behaviour and violence against them in the context of family and/or intimate relationships with a spouse. This finding is supported by a body of literature showing that violence against women leads to poor mental health outcomes and ultimately suicidal behaviours especially among women of ethnic minority and/or migrant backgrounds (Abramsky et al., 2011; Bhugra et al., 1999a; 1999b; Devries et al., 2013; Colucci & Heredia-Montesinos, 2013; Canetto, 2015). In some studies, mental illness (notably depression) were more commonly reported among females compared to males or ethnic majority females. This relates to the work of Devries and colleagues (2013), who provided meta-analytical evidence of the predictive role of the intimate partner violence on depression symptoms and attempted suicide among women worldwide (Devries et al., 2013). Arguably, the frequency of the mental and physical illnesses in this current review points to the lack of physical, sexual and emotional safety. Specifically, impaired autonomy, economic dependence and poverty render women particularly vulnerable to distress, which is a tremendous setback to mental health equity (Douki et al., 2007; Abramsky et al., 2011; Devries et al., 2013; Canetto, 2015). An interesting finding is that women living in traditional areas of 168 ― Chapter 5

Turkey reported less mental illness than those living in less traditional areas. However, the underreporting of mental illness due to stigma in traditional areas may explain this result (Eylem et al., 2016). Alternatively, the lack of mental health professionals in traditional regions, who could provide a diagnosis, could also explain it. On the other hand, psychiatric diagnoses of individuals who died by or attempted suicide may reflect a tendency of Western scientists to medicalise social and political problems, including violence, while the problem of high suicidality should be combatted as a form of social oppression (Usser, 2010). Thus it is relevant to ask whether the gender related factors such as “honour-related” violence rather than mental illnesses are more relevant to the suicidality of Turkish females in traditional areas? With regards to the preceding question, “honour-related” violence was often speculative in the included studies from Turkey. Even though the findings of the studies from Europe (e.g. Heredia Montesinos et al., 2018; van Bergen et al., 2009) were more robust and empirically tested, the scale of the studies was often smaller. Contrary to the results of this review, one large scale community study (N= 414) of women aged 13 and 54, provided extensive evidence on severe forms of autonomy restrictions and “honour- related” violence in traditional rural areas in Turkey (KAMER, 2011). Accordingly, 55.8 % women aged below 20 were married through an arranged marriage and 24.5% were forced into a marriage. Further, all types of violence were common and percentages included; 99.3% psychological violence, 89.4% physical violence, 91.3 % verbal violence, 90.8% economic violence and 63% sexual violence (KAMER, 2011). These differences in findings of the included studies and the former community study point to the possibility that “honour-related” violence specifically is not sufficiently investigated in the scientific literature. It could also be that within “domestic violence” and “family problems” categories of the included studies, ‘honour’ actually played a role as well. Thus, it remains unclear whether there is a variation in “honour-related” violence depending on the region of the study. Another interesting finding was related with religiosity. Some studies indicated that religiosity was lower among female suicide attempters in traditional areas of Turkey Attempted Suicide and Suicide ― 169

(Özdel et al., 2009; Yaşan, 2008). This finding suggests that some Turkish women lacked religion to protect them from suicidal behaviour (Lawrence, Oquendo & Stanley, 2014). Alternatively, Turkish women who are not religious could experience a sense of isolation since they are surrounded by mostly strict believers in highly traditional areas. The latter argument is supported by a recent multinational study on the role of religion in suicidal behaviour and attitudes by Eskin and colleagues (Eskin et al., 2019). In Muslim countries, where the freedom of religion is restricted or religion is compulsory, the protective function of the religion is arguably limited (Canetto, 2015; Eskin et al., 2019). Taken all together, the results of the current study could be interpreted within the framework of intersectionality, in which broader structural inequalities in gender role expectations (e.g. self-sacrifice of woman for the family) and power imbalances in social class (e.g. low socio-economic status of women) collectively impact on mental health 5 outcomes (e.g. depression) and suicidal behaviours of Turkish women (Baker, Procter & Ferguson, 2016; Canetto, 2015; Rezaeian, 2010; Hjelmeland & Knizek, 2017). Arguably, the consistency in socio-demographic characteristics emphasises the fact that the specific sub-group of Turkish women aged 16 and 24 often occupy the lowest social status in highly traditional Turkish communities when compared to men and older women (Eldering, 2014). Furthermore, the gender related precipitants, remind the key features of Durkheim`s fatalistic suicide construct including, lack of control over one’s life course and lack of personal autonomy (Durkheim, 1897, 1952). Thus, the suicidality of this specific sub- group could be understood as an attempt to communicate their sense of powerlessness and alienation in response to the harsh moral demands that are upheld through force (Canetto, 2015; van Bergen, 2009; Schouler-Ocak, 2015; Aichberger et al. 2015; Bagli & Sever, 2003). Additionally, drawing on the interpersonal theory of suicide developed by Joiner (Joiner, 2005), it could be argued that suicide attempt and suicide are attempts to escape from the sense of powerlessness and abuse in the only ways that seem possible (Selby & Joiner, 2009). 170 ― Chapter 5

Strengths and Limitations An important asset of this study is that it is systematic, in terms of the search strategy, inclusion, and in terms of the information extracted and reported. It also reports on the quality of the included studies. Yet another strength of this review is that it draws on the literature published in four languages (Turkish, Dutch, German, English) and synthesises knowledge on suicidality of Turkish women from the Turkish and the European contexts. Nevertheless, this study also has limitations. With respect to the quality of the studies, some studies did not consistently present gender differences or failed to present percentages when they did investigate this topic. A majority of the studies investigating attempted suicide collected data retrospectively and relied on a single source of information (e.g. hospital records). Even when they used mixed methods (direct interviews with informants and hospital records), they did not describe any procedures to deal with the discrepancies in information drawn from different sources. Another issue was that since the majority of studies of attempted suicide were conducted very shortly after women were admitted as patients to a hospital, reports by hospital staff were much more common than self-report. These flaws or omissions in methodology or the presentation of data indicate that the conclusions of some of the studies may be biased (Conner et al., 2012). Further, studies rarely used controls. Without a comparison group, the uniqueness of certain characteristics and/or precipitants remains unclear. Additionally, prospective studies investigating attempted suicide were scarce. Such a paucity prevents a solid check for support or care for women after an attempt or the monitoring of possible variations in precipitating factors within cases over time. (Hjelmeland & Knizek, 2017). In fact, one prospective study pointed to the increased violence against female suicide attempters after their attempt in a traditional area in Turkey (Yaşan, 2008). One of the major limitations of the study is the possible variability of data collection, reporting and selection of the analysed factors across countries and Turkish regions. As briefly mentioned in the beginning of our discussion, honour-related violence for instance, was not sufficiently empirically investigated in the studies from Turkey, and only one study from Europe assessed discrimination and stigma against immigrants. Attempted Suicide and Suicide ― 171

This means that the role of migration and belonging to an ethnic immigrant minority group in Europe for understanding suicidal behaviour of Turkish women remains inconclusive.

Implications and Conclusions To our knowledge, this is the first systematic literature review comparing the characteristics and precipitating factors associated with suicidal behaviours in Turkish women in Europe and Turkey, and its results point to a consistency between the two geographies. Future research should systematically address gender differences in the investigation of precipitating factors and characteristics, include a representative control group and make more use of prospective designs. Furthermore, future studies could investigate whether the risk and precipitating factors of suicide among women are 5 influenced by traditional culture and patriarchal systems, and they could explicitly examine the harmful role of honour. For the latter, qualitative or mixed methods designs may be appropriate. To summarise, suicidality everywhere is related to a complex interaction of cultural, social, economic and individual factors which are varied depending on gender, region and socio-political contexts (Canetto, 2015; Hjelmeland & Knizek, 2017). Turkish women`s suicidality can be interpreted as an attempt to communicate their discontent in response to social injustices and/or an attempt to escape from these injustices, which they are exposed to in the name of honour. These are clear examples of serious human rights violations (Canetto, 2015). There is an empirical evidence from the Netherlands indicating that cultural change in such gender specific “honour” codes (i.e. restraint in behaviour such as modesty) are possible through psycho-educational training programmes among ethnic groups with high orientation on “honour” (Turkish, Moroccan) compared to those with low orientation on “honour” (Dutch) (Cihangir, 2013). To conclude, suicide prevention efforts in this specific risk group should not only target reducing suicidal behaviours but should also address cultural beliefs and attitudes underlying gender related precipitants through such education programmes. In addition to education, 172 ― Chapter 5 culturally sensitive domestic violence services, early intervention and/or detection of young women who might be at risk are important. See for instance, the suicide awareness campaign “End Your Silence Not Your Life” targeting Turkish immigrant women in Berlin (Schouler-Ocak, 2015). Finally, adequate further jurisdiction and political action to prevent human right violations against women in Europe and Turkey is much needed in order to prevent suicidal behaviours (Lester, 2014) Attempted Suicide and Suicide ― 173

5

6

protocol: redUcing sUicidal ideation among tUrkish migrants in the netherlands and in the Uk: effectiveness of an online intervention

pUblished as Eylem, O.,van Straten, A., Bhui, K., Kerkhof, A.J.F.M (2015). Study protocol: Reducing suicidal thinking among Turkish migrants living in The Netherlands and in the UK: Effectiveness of an online intervention.International Review of Psychiatry,1,72-81. 176 ― Chapter 6

abstract

Background The Turkish community, living in Europe, has an increased risk for suicidal ideation and attempted suicide. Online self-help may be an effective way of engagement with this community. This study evaluates the effectiveness of a culturally adapted, guided, CBT based online self-help intervention targeting suicidal ideation for Turkish adults living in the Netherlands and in the UK.

Methods This study will be performed in 2 phases. First, the Dutch online intervention will be adapted to Turkish culture. The second phase is a randomized controlled trial with 2 conditions: experimental and waiting list control. Ethical approval has been granted the trial in London and Amsterdam. The experimental group obtains direct access to the intervention, which takes 6 weeks to complete. Participants in the waiting list condition obtain access to the modules after 6 weeks. Participants in both conditions are assessed at baseline, post-test and followed up 3 months after post-test. Primary outcome measure is reduction in frequency and intensity of suicidal thoughts. Secondary outcome measures are self-harm, attempted suicide, suicide ideation attributes, depression, hopelessness, anxiety, quality of life, worrying and satisfaction with the treatment.

Conclusions This is the first self-help intervention that is adapted to Turkish adults presenting with suicidal thoughts. This will bring further efforts to identify vulnerable and yet underrepresented ethnic groups. Protocol ― 177

6 178 ― Chapter 6

introdUction

uicide is a major global concern for public health. It has enormous emotional, social and economical implications at the individual as well as collective levels (Bertolote & Freischmann, 2009; van Spijker et al., 2011). The definition of suicide is still fiercely debated. One approach suggests San overlap between behaviours (e.g. suicidal ideation, attempted suicide, self-harm) that together constitute the suicide spectrum (De Leo et al., 2006; Lawrie et al., 2000). Prospective studies suggest that people with suicidal ideation are three times more likely to proceed to self-harm compared to people who do not have such thoughts (e.g. Fergusson & Lyskey, 1995; Fergusson et al., 2005). The continuum between suicidal thoughts, worries and behaviours may suggest that people who are worried with suicidal thoughts are more likely to engage in self-harm and suicide attempt (Kerkhof et al., 2011). There is an increased likelihood of repeating the self-harm as a consequence of habituation and there is a greater risk of death by suicide (Joiner, 2005). Knowledge about suicide in diverse cultural group is limited by theories generated from research among majority communities living in Western countries. Belonging to an ethnic minority however, has been considered as one of the risk factors leading to suicidal behaviours (WMR, 2010). The lower utilization of mental health services among minorities is a concerning public health issue (Lindert et al., 2008; Bhui et al., 2003; van der Stuyft et al., 1989; Juang & Cookston, 2009; Bhugra, 2004). The low up-take of mental health services is thought to reflect cultural and linguistic barriers during the help-seeking process. There is a need for more knowledge about high-risk groups including ethnic minorities, and about culturally specific conceptualizations of suicide, as well as help-seeking and pathway to receive care (Lester, 2012; Canetto, 2008; Bhui, 2010; Hjelmeland, 2011). In the current literature about suicide, there is insufficient information on overall situation of migrants in Europe. More is known about the suicidal behaviours among some ethnic groups with a long history of migration in Europe such as South Asians Protocol ― 179 in the UK (e.g. Bhugra et al., 1999a; Bhugra et al., 1999b; Bhugra, 2004; Bhugra, 2002; Hunt et al., 2003; Bhui et al., 2007; Patel & Gaw, 1996) and less is known about other ethnic groups which are becoming prominent ethnic minority populations in Western Europe such as the Turkish speaking community (Burger et al., 2009; van Bergen et al., 2008; van Bergen et al., 2009; van Bergen et al., 2010; Yılmaz & Riecher-Rössler, 2012; Razum & Zeeb, 2004). The Turkish speaking community in Europe is one of the fastest growing communities that consists of culturally and linguistically diverse ethnic groups (e.g. Kurdish people from South East region, Laz from Black Sea region). Their history of migration starts with the labour migration agreements in the early 1960s with Western European countries, most notably with Germany. The outflow of Turkish migrants continued mostly to England for various reasons (e.g. labour, political regression) and to Germany for family reunification in the 1970s (Taş et al., 2008; Kilberg, 2014). Today it is estimated that more than 5 million people of Turkish origin live abroad (Kilberg, 6 2014). In Europe, the largest Turkish migrant populations live in Germany (18.5 % of the German population), France (7 % of the French population) and in the Netherlands (2.3% of the Dutch population) (Kilberg, 2014). It is difficult to establish the accurate statistics in the UK due to absence of a separate category for Turkish people in national statistics (Taş et al., 2008). The available literature indicates that the Turkish speaking community, living in Europe, faces special risks for attempted suicide (e.g. Lindert et al., 2008; Burger et al., 2009). In the Netherlands, people of Turkish origin die because of suicide at a younger age compared with indigenous Dutch people (Turkish male: 32, Turkish female: 32; Dutch male: 48, Dutch female: 51) (Garssen & Kerkhof, 2006). There is also an increased risk of suicidal ideation (38.1% for Turkish adolescents vs 28.9% for Moroccan and 17.9% for Dutch; van Bergen et al., 2008). Furthermore, young Turkish females (15-24) have a 1.6 times higher risk of attempted suicide compared to young females (15-24) from other ethnic backgrounds (Moroccan, Surinamese and Dutch) even though the risk of completed suicide was not increased (Burger et al., 2009). 180 ― Chapter 6

High levels of attempted suicide but a lower suicide mortality rate among young women of Turkish origin was found in Germany and Switzerland compared with indigenous populations (Razum et al.,1998; Razum & Zeeb, 2004; Yilmaz & Riecher- Rossler, 2008; Yilmaz and Riecher-Rossler, 2012). In Turkey, young women are at higher risk for suicide (Devrimci-Özguven and Sayıl, 2003 ; Polatöz et al., 2011; Bağlı & Sever, 2003; Çakmak & Altuntaş, 2009; Clemens, 2011). Overall, in the Turkish community, the most common reasons for suicide are reported as social isolation, intergenerational conflict or problems in relationships (Razum & Zeeb, 2004; Yılmaz & Riecher-Rössler, 2012). In a case file study, it was identified that at least in half of the cases, Turkish women experienced stressful life events related to family “honour”, questioning of cultural values and self-sacrifice (van Bergen et al., 2010). One of the possible explanations for this higher risk for suicidal behaviours in the Turkish community comes from the level of acculturation. Acculturation is defined as a process that includes exchange of cultures (Berry & Sabatier, 2011; Roccas et al., 2000). Different acculturation strategies (i.e. separation, integration, marginalization and assimilation) differentially relate to mental health outcomes such as depression and suicidal ideation across ethnic groups. Prospective and cross-sectional research evidence suggests that the cultural integration strategy (i.e. positive attitudes towards the heritage culture and the receiving culture) is associated with a lower risk for psychological distress, depression symptoms and suicidal thoughts compared to other acculturation strategies such as separation (i.e. positive attitudes towards the heritage culture and negative attitudes towards the receiving culture) and marginalization (i.e. negative attitudes towards both cultures); these findings are relevant to a number of ethnic communities (e.g. South Asians, Moroccans, Turks, Black Caribbean, Black African; Bhui et al, 2012; Juang & Cookston, 2009; Fassaert et al., 2009; Ünlu İnce et al., 2014; van Leeuwen et al., 2010; van der Stuyft, 1989). Although the choice of a particular acculturation strategy seems to provide a plausible explanation for the risk for suicide, it is not sufficient to explain why this risk is higher in Turkish speaking communities compared to the other ethnic minority communities sharing migration experiences. It also does not explain why there is a consistency Protocol ― 181 between high attempted suicide risk among young Turkish women in Turkey as well as in Europe. The findings suggest pathogenic effects of culture. From a sociological perspective, one can argue that in the Turkish community suicidal behaviours are related to the culturally specific conceptualization of gender roles transmitted through the generations ( e.g. White, 1997; Diehl et al., 2009; Çakmak & Altuntaş, 2008; Lipsitz Bem, 1981 ). Thus, a patriarchal family structure, where men are responsible for the “namus” (i.e. honour) of the women in the family, bring bring restrictions (e.g. not being allowed to go out without a companion of male relative) in the life of women of Turkish origin (e.g. van Bergen et al., 2010; Diehl et al., 2009). Questioning of such restrictions is likely to result in tension between women and the rest of the family. This is perhaps perceived as a “threat” to the unity and the reputation of the family in the community and followed by more tension and further restrictions to control women`s behaviour. In this context, suicide might seem as the only way to escape from this coercive pattern and the associated feelings (Baumeister, 1990). 6 Online psychological treatments have recently become an addition to the mainstream health system. Given their effectiveness, especially in reducing anxiety and depression symptoms (e.g. van`t Hof et al., 2009), they are promising intervention for improving public mental health. Recent developments suggest that online CBT- based self-help is also effective in reducing suicidal thoughts in the mainstream Dutch population (van Spijker et al., 2014). Although the effect sizes were small, the intervention group showed reduction in suicidal thoughts (d=0.28), hopelessness (d=0.28), worry (d=0.34) and improved health status (d=0.26). These results remained in three months follow-up study. Online psychological services have several advantages compared to face-to-face treatment. For example, it is relatively easy to offer treatments in different languages on the internet. Such a programme offers flexible services which might be beneficial for example for Turkish people who don`t seek help because of their fear of being judged or misunderstood by the care-provider (Taş et al., 2008). Additionally, in the Netherlands, it is estimated that almost 80 % of the Turkish-speaking population have access to the internet. The younger generation, which is at greater risk for suicide, has more 182 ― Chapter 6 frequent access than the older generation (CBS, 2012). Further evidence, that supports the value of online interventions comes, from a trial by Unlu and colleagues who tested the effectiveness of a self-help intervention, which was tailored according to the Turkish culture, in reducing the depression symptoms in the Netherlands (Ünlü İnce et al., 2013). The intervention group showed significantly more improvement in post-test (6 weeks; d=1.68) and in follow up (3 months; d=1.13) compared to the waiting-list control group. Although this study was underpowered, culturally adapted online services seem promising for this particular population (Ünlü İnce at al., 2013 ; Ünlü İncet al., 2014c). In summary, there is a need to further our understanding of suicidal behaviours particularly for ethnic groups such as the Turkish speaking community which is currently under-represented in health services and in official statistics on service-use. In the current study, acculturation is considered as a potential co-variate to explain the direction of the relationship between the utilization of online modules and reduction in suicide ideation and in other outcome measures such as depression scores (Bhui et al, 2012; Juang & Cookston, 2009; Fassaert et al., 2009; Ünlü İnce et al., 2014c; van Leeuwen et al., 2010; van der Stuyft, 1989). Furthermore, self-harm is included as one of the outcome measures in keeping with the evidence suggesting self-harm as one of the behaviours comprising the suicide spectrum (e.g. Hawton et al.1999; Fergusson & Lyskey, 1995; Fergusson et al., 2005). Given the complexity of the background leading to suicidal behaviours in the Turkish speaking community (e.g. Lindert et al., 2008; Polatöz et al., 2011; Çakmak and Altuntaş, 2009; Akpinar, 2003; van Bergen et al., 2010), it is important to make these psychological interventions relevant to their presentations of mental health symptoms (Hjelmeland, 2011). Online forms of service-delivery are promising ways of accessing socially excluded populations and other ethnic minorities who need services but cannot access them easily. In light of these, the overarching objective of this study is to adapt, and to test the effectiveness of CBT based modules for Turkish speaking population. Protocol ― 183

methods/ design This study is performed in two phases. In the 1st phase the online intervention, which is used in the work of van Spijker et al (2012), will be translated and adapted for the Turkish population. The 2nd phase is a RCT comparing a web-based self-help intervention with a waitlist control condition.

Adaptation Procedure The adaptation of the intervention will be based on focus groups with lay members of the Turkish community and one-to-one interviews with the CBT therapists and mental health professionals working with Turkish communities. Focus groups were chosen as a method to explore how Turkish people construct the meaning of suicidal behaviours, and how they describe suicidal behaviour and remedies for it (e.g.Chu et al.,2012; Rathod et al.,2010; Feingber et al., 2012). In a group setting, interactions between individuals who share a similar cultural heritage can bring out diverse perspectives that we will 6 need to accommodate in the material of the intervention and in information sheets given to patients when seeking consent. So focus groups will permit access to a range of perspective all identified within Turkish culture and will make such perspectives amenable to the epidemiological part of this study (Kohrt et al., 2013). One-to-one interviews will be conducted to learn about professional conceptualization of suicidal thoughts and helpful and/or unhelpful interventions while working with the Turkish speaking community. The information from the focus groups and the professionals will be used to modify the language content, for example, specific words used to describe distress; and to tailor the assignments as presented in the intervention. The core of the intervention, i.e. CBT, will not be changed. The aim is to interview 6 professionals and run 4 focus groups with a maximum of 6 participants in each group. The focus groups and the interviews with professionals will be in Turkish and/or in English or Dutch. Participants aged 18 or older with Turkish origin, who have some experience with using mental health services, are eligible to 184 ― Chapter 6 participate in groups. For the purpose of this research, the Turkish culture is defined on the basis of the language and a place of birth of either the participants or one of their parents. Thus, participants who were born in Turkey or have at least one parent born in Turkey are included. Participants will be recruited through the community (banners on websites, social media, flyers, etc). All interested people will receive an information letter about the goal of the focus group, the way it will be performed, and the way data is handled.

Statistical Analysis of the interviews Thematic analysis will be used as a means of identifying, analysing and reporting explanatory models, and for understanding how elements of the intervention might need modification; patterns (themes) within the recipients’ and professionals’ descriptions of their experience of suicide, seeking help or disclosing suicidal behaviours. This method minimally organises and describes the data set in rich detail. However, it frequently goes further than this, and interprets various aspects of the research topic (Braun & Clarke, 2006).

RCT studying the effectiveness of the intervention The second phase of the study is RCT with two conditions: online intervention and waitlist control.

Sample size The sample size of this study is based on the expected effect of the primary outcome measure: the reduction in frequency and intensity of suicidal thoughts as measured with the BSS. We expect an effect size of d=0.40. In the previous trial, in the Dutch population, an effect size of 0.28 was found (van Spijker, 2012). There is however, suggestive evidence from trials indicating larger effect sizes for guided cognitive behavioural based interventions compared to unguided ones (e.g.Cuijpers et al.,2010). Based on power of Protocol ― 185

0.80 and an alpha of 0.05, 100 participants are needed in each condition. Given the expected drop-out of 30%, the total sample size is determined as being 286.

Inclusion and exclusion criteria In order to be eligible to participate in this study, a participant must meet all of the following criteria: 18 years and older, being Turkish, which is defined as being born in Turkey or having at least one parent being born in Turkey and speaking the Turkish, Dutch or English language, and presenting mild to severe suicidal ideation. This is defined as a score 1 and above on the Beck Suicide Severity (BSS) scale. Total score of the BSS ranges from 0 to 38, with higher scores indicating more severe suicidal thoughts. There is no empirical evidence to support the use of a specific upper cut off score with the BSS (Beck & Steer, 1993). There is no previous study demonstrating the use of specific BSS cut off scores in Turkish migrant population. Thus, no upper cut off score will be used in order to reach the upper end of the suicidal ideation spectrum. Participants also 6 need to have an access to PC and internet and have sufficient command of English, Dutch or Turkish. Finally they need to be willing to provide their name, telephone number and e-mail address of themselves as well as from their GP.

Inclusion Procedure Recruitment will take place among Turkish speaking population living in the Netherlands and in the UK. Participants will be recruited from the general population through banners on relevant websites (e.g. of community and health organisations), and through social media. A study website will be created including information about the study where potential participants can register themselves. Those who do will receive an e-mail with further information, an informed consent form and a link to the baseline questionnaire. Those who return the informed consent, provide their contact details and those of their GP, fill out the baseline questionnaire, and do not fulfil our exclusion criteria will be included. 186 ― Chapter 6

Randomization The randomization scheme will be derived using random allocation software by an independent researcher. Randomization will take place in blocks of 20, and will be stratified for the UK and NL. Randomization will take place in a 1:1 ratio. All participants will be informed about the condition they are assigned to.

Ethical Implications The study has been reviewed by the ethics committees in the UK as well as in the Netherlands. One of the major ethical issues was related with the safety of the participants in both conditions in the study. In the previous trial run by van Spijker and colleagues, none of the participants in both conditions died because of suicide during the study (van Spijker et al., 2012). Thus, in the current study we decided to follow the same safety procedure in both conditions. Given the language barrier, we will provide the resources and information about how to seek help in Turkish in two countries.

Safety of participants The safety protocol, which is previously developed and followed by van Spijker, will be followed in order to ensure the safety of participants (van Spijker, 2012). This means that suicidal thoughts will be assessed once in every 2 weeks. If a participant scores above the cut-off scores (BSS > 29) then the researcher will ring the participant to do a risk assessment. In case the participant does not answer his / her phone, he or she will be called for 3 working days on different times of the days. In case of no response after 3 days of attempting to call, a standardized e-mail will be sent asking the participant to contact us. In addition, in case we cannot reach the participant we will contact the GP to inform him / her about the high suicide ideation. This will also be mentioned in the e-mail to the respondent. When we do have contact with the participant we will assess the suicide risk and contact the GP if necessary. This whole procedure is explained to potential participants in the information letter. In addition it should be noted that Protocol ― 187 participants in both conditions are encouraged to access treatment while they are enrolled in the programme. They will be provided with links to relevant health care providers.

Intervention The online self-help programme consisting of 6 modules, which is developed by van Spijker and colleagues, will be used as the core of the program. The intervention is described in detail by van Spijker and colleagues (see van Spijker, 2012). For the current study, we added two modules about self-harm in keeping with the evidence suggesting an overlap between suicidal worries and thoughts leading to attempted and completed suicide (e.g. Hawton & Fagg, 1988; Hawton et al., 2003; Fergusson et al., 2005). We encourage people to choose 6 out of the 8 modules so that the intervention can still be finished within 6 weeks. Every module consists of information and exercises. Ideally, participants need to spend 15 minutes twice a day to perform the exercises. 6

The content of the eight modules is: 1. Thinking about suicide The aim of this week is to teach the participant to gain some control over their suicidal thoughts and to identify their automatic thoughts that underlie their suicidal thoughts. In addition, attention is given to the dichotomous and overgeneralising character of the suicidal cognitions. 2. Dealing with thoughts and feelings Participants will learn to realize that seemingly unbearable feelings can be tolerated. Attention will be given to learning to tolerate and regulate intense emotions in a crisis situation. 3. Thinking about the future Participants will work on their ideas of the future. Are these realistic? Subsequently, participants will compose a realistic idea of the future and formulate new goals. 4. Thinking about the self Common mistakes in thinking regarding the self are discussed and worked on. In addition, it is explained that suicidality is a long-term vulnerability. 188 ― Chapter 6

The participants will be taught how to handle this. Also attention is given to learning to ask for help. 5. Thinking about others Mistakes in thinking regarding other people will be discussed. In addition, information about the consequences of suicide for relatives and friends will be reviewed. 6. Repetition and relapse Several important mistakes in thinking are repeated, as well as several learned skills. Furthermore, attention will be given to relapse and how to prevent this. 7. Acknowledging self-harm Participants will gain an understanding about the role of self-harm in their lives. They identify the bad as well as the good things about the behaviour. Exercises focus on naming the behavior (e.g. cut) in order to help participants to think of their behavior without feeling embarrassed or ashamed about it, which are common barriers in disclosing the behaviour. 8. Setting Targets At this part, participants are encouraged to set realistic targets about their future by using motivational interviewing techniques and future oriented reflexive approach (e.g. how does self-harm effect your life? If things were to improve, what might you find yourself doing?).

Guidance Guidance is provided by moderators (2 moderators in the UK and 2 moderators in the Netherlands), who are supervised by the researcher. They need to be bi-lingual (English-Turkish or Turkish-Dutch) psychology students at a Masters level. Series of meetings will be arranged with them where they will receive training about how the online intervention works, safety protocol, referral system in both countries and their roles and responsibilities whilst providing guidance to participants. Following their agreement with regards to their role as a moderator, regular supervision meetings, which will be held through skype and/or e-mail communications will be arranged with them. After each lesson the participants are requested to do ‘homework’. They will receive online feedback after the completion of this homework. The aim of this personalized Protocol ― 189 feedback is to help participants understand the methods as explained in the lessons. Moreover, it is used to motivate the participants to continue with the programme. In case the participant does not send in homework, he or she will be contacted by the moderator to see if the participant is struggling working on the programme.

Waiting list Control Condition Participants in the control condition will have access to a website where information about suicide is provided. In addition, specific information about treatment options (e.g. 113 online, Samaritans, mental health organisations) will be provided. After six weeks, after the post-test questionnaire, participants can start with the intervention.

Instruments All assessments are self-report instruments and will take place online in the participant’s own preferred language, either, Turkish, English or Dutch. Assessments will take place 6 before the randomization (t0), after completing the treatment (t1; 6 weeks) and 3 months after completing the treatment (t2).

Primary Outcome Measure The primary outcome measure in this study is the reduction in the frequency and intensity of suicidal thoughts. This will be measured with the Beck Scale for Suicidal Ideation, self-report (BSS) (Beck et al.,1988). The BSS is a 21-item measure assessing the severity of the suicidal ideation (Beck et al., 1974; Beck et al.,1979). Each item is scored from 0 to 2. The total score is obtained by adding the first 19 items and it ranges from 0 to 38. High score represents high suicidal ideation. The BSS has good psychometric properties in English (Beck et al., 1988, Beck et al., 1979) and Turkish (Dilbaz et al., 1995; Polatöz et al., 2011). 190 ― Chapter 6

Secondary Outcome Measures Secondary outcome measures include: depression, suicidal ideation attributes, hopelessness, anxiety, worrying, quality of life, self-harm behaviour, suicide attempt, acculturation, and satisfaction with the treatment. Depression will be measured with the Beck Depression Inventory (BDI) comprised of 21 items(Beck et al., 1961). Each item is scored from 0 to 3. The severity ranges from minimal depressed (0-13) to severely depressed, indicated by a score of 29 and over (Beck et al., 1961). It is reliable and valid (Beck et al., 1961). The BDI has been validated in Turkish and Dutch population (Hisli, 1987). Suicidal Ideation Attributes The Suicidal Ideation Attributes Scale (SIDAS) is a five-item scale assessing frequency (item 1), controllability (item 2), closeness to attempt (item 3), distress (item 4) and interference with daily activities (item 5) on 10-point scales over the past month (van Spijker et al., 2014). Total SIDAS scores are calculated as the sum of the five items with total scale scores ranging from 0-50. Higher scores are indicative of greater suicidal ideation severity. An initial validation study of this scale demonstrated that the scale has high internal consistency and good convergent validity (van Spijker et al., 2014) Hopelessness is measured with the Beck Hopelessness Scale (BHS), which contains 20 true and false statements (Beck and Steer, 1988). Each statement is scored from 0 to 1 and the total score ranges from 0 to 20. A high score indicates high degree of hopelessness. The instrument has good psychometric properties (Brown, 2011). This is also true for the Turkish version (Seber et al., 1993). Anxiety is measured with the Anxiety subscale of the Hospital Anxiety and Depression Scale (HADS). (Zigmond & Snaith, 1983; Spinhoven et al., 1997). The anxiety subscale consists of 7 items. Each item is scored on a 4 point Likert scale with scores ranging from 0 (no anxiety) to 3 (high anxiety). Total score range is 0-21 (Spinhoven et al., 1997). The scores between 0 to 7 indicate no anxiety, scores between 8 to 10 indicate possible anxiety and scores above 11 or 12 are indicative of severe clinical anxiety. The HADS has been demonstrated as a valid and reliable instrument in both Protocol ― 191 various Dutch and Turkish populations (Spinhoven et al., 1997; Bjelland et al., 2002; Aydemir et al., 1997). Worrying The Penn State Worry Questionnaire (PSWQ-PW) is a 15 item inventory assessing both the weekly status of pathological worry and treatment related changes of worry during the treatment (Stöbber & Bittencourt, 1998). Each item is scored on a 7 point rating scale, ranging from never 0 (never) to 6 (almost always). The total score ranges from 0 to 90 with a high score indicating more worrying. PSWQ-PW shows good reliability and convergent validity (Stöber and Bittencourt, 1998). The Turkish version demonstrated reliability (Yılmaz et al., 2008). Quality of life The Euro Quol (EQ-5D) is an instrument measuring health quality of life and comprised 5 items: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression (Rabin and Charro, 2001). Each item is required to be rated as 1 (no problem), 2 (some problem) or 3 (extreme problem). The current health state is also rated on a scale ranging from 0 (worst imaginable state) to 100 (best imaginable state). Both 6 Dutch and Turkish versions have been validated (Lamers et al., 2005; Eser et al., 2007). Self-Harm and Suicide Attempt A question asking about a suicide attempt and will be used to assess the presence of a suicide attempt during the course of the study. Additionally the 3 screening questions measuring the presence of self-harm will be taken to use from the original instrument (Eylem, 2011). These questions are chosen to be included as they are formulated in line with the evidence supporting the overlap between suicidal thinking, self-harm and suicide attempt (e.g. Hawton et al., 1999; Fergusson et al., 2005). Additionally, initial findings provide supportive evidence for the concurrent validity of the tool (Eylem, 2011). The rest of the questionnaire include 12 items assessing specific methods, frequency, reasons, severity, feelings before and feelings after self-harm, specific motivations for self-harm, drug or alcohol use and communication before and after self-harm. Further studies regarding the psychometric properties of this tool are in progress. Satisfaction with the Treatment Participants will be asked to define their satisfaction with each lesson by asking “Was this lesson useful to you?” in Dutch, 192 ― Chapter 6

English and Turkish. The answers will be rated on a 5-point Likert scale. The score per item ranged from 1 (not at all) to 5 (very much).

Covariates

Acculturation is measured with the Lowlands Acculturation Scale (LAS) measures the degree of acculturation (Mooren, 2001). Acculturation is defined as a dynamic process which includes exchange of cultures (Berry and Sabatier, 2011; Roccas et al., 2000). It consists of 25 items which are rated on a 6 point scale (0: totally disagree, 6: totally agree). LAS is divided into 5 subscales: skills, tradition, social integration, values and norms and feelings of loss. The validated Turkish, Dutch and English versions of the instrument are available (Mooren, 2001).

Statistical analyses of the RCT The study will be carried out in accordance with the CONSORT guidelines. All analysis will be based on the intention-to-treat sample and missing values will be imputed with multiple imputation procedure as implemented in SPSS. ANOVA will be used to compare the post-test mean scores for the intervention group with the post test scores for the control group. Cohen’s d will be used to determine the effect size of the differences in post-test means. Cohen`s d will be calculated as the difference between the post t-test mean scores of the intervention and control group divided by the pooled standard deviation. Effect sizes of 0.8 are accepted as large, effect sizes of 0.5 are moderate and effect sizes of 0.2 are small (Cohen,1988).

Summary and Implications The current paper describes a study protocol of a randomized controlled trial comparing a web-based intervention, which is adapted to the Turkish culture, to reduce suicidal thinking with a control condition. To our knowledge, this is the first self-help intervention that is adapted to Turkish adults presenting with suicidal thoughts. This will perhaps bring further efforts to identify vulnerable and yet underrepresented ethnic groups. Protocol ― 193

It is also equally important to use this knowledge to develop prevention programmes and to promote policies that are inclusive and encourage access to effective services. Another highlight of this study is the interest to explore the role of acculturation (i.e. whether greater participation in Dutch and British society is related with the increased engagement with the online modules and reduced suicidal thinking) in moderating the relationship between for example use of online modules and reduced suicidal thoughts. This will potentially further our understanding of suicidality in a particular vulnerable ethnic group in Europe. In the current study, methods from the formative model of adaptation such as focus groups with the target population and one-to-one interviews with the professionals working with them will be used (Hwang, 2009). Similar methods of adaptations have been tested and demonstrated high participant satisfaction, adherence and retention (e.g. Rathod et al., 2010; Feingber et al., 2012; Chu et al., 2012; D`angelo et al., 2009). Given the diversity within Turkish culture itself, simply translating the Dutch online 6 modules into Turkish language will not be sufficient to make the intervention appealing to participants (Bernal et al., 1995; Bernal, Jimenez-Chafey & Rodrigez, 2009; Kohn et al., 2002). In light of these, the design of this study will contribute to the development of a research process which is robust to international contexts and can be applied to Turkish people in any country. This will make the current research more generalizable and of value. There will be lessons for working with ethnic minorities in general also, where the diaspora experience different country contexts. There are other strengths about the design of this study. First, running the trial in two countries will make it more likely that sufficient number of people will be recruited in both conditions. The power was an issue in previous trial given the exclusion of people presenting with suicidal thinking, high drop-out rate and difficulty with recruitment in this population (Ünlü et al., 2013; Ünlü et el., 2014). Another important strength is using moderators, representing the Turkish community, whilst engaging and collecting data. It is expected that, their involvement will reduce the attrition rate and ensure the safety of participants. 194 ― Chapter 6

There are some limitations. In spite of the use of guidance, drop-out attrition happens in internet-based self-help interventions (Eyesenbach, 2005). This can potentially introduce a selection bias which limits the conclusions drawn from the study. Second, given the sensitivity of the study, anonymous participation is not possible. This might also introduce a selection bias. Third, no diagnostic interview is conducted due to the broad inclusion criteria, time limitation and the self-help character of the study. Lastly, although the validated instruments, which are available in 3 languages, have been used, their psychometric properties in an online environment are not sufficiently known. To conclude, to our knowledge, this study has several unique aspects which will contribute to the literature about cultural adaptation and suicidality. These are: targeting a specific ethnic group which is vulnerable and yet under-represented in the literature as well as in the mainstream health care-system, being a cross-cultural study and using moderators from the community organisations. Protocol ― 195

6

7

redUcing sUicidal ideation among tUrkish migrants in the netherlands and in the Uk:a feasibility and pilot rct of an online intervention

sUbmitted as Eylem, O., van Straten, A., de Wit, L., Rathod, S., Bhui, K., Kerkhof, A.J.F.M, (2020). Reducing Suicidal Ideation among Turkish Migrants in the Netherlands and in the UK: A Feasibility and Pilot RCT of an Online Intervention 198 ― Chapter 7

abstract

Background There is growing evidence for the effectiveness of e-mental health interventions in reducing suicidal ideation. The evidence for the effectiveness of e-mental health interventions among ethnic minorities is still preliminary. This mixed methods study investigates the feasibility of a culturally adapted, guided online intervention with the intention to understand how it works and for whom to inform refinement. It also examines indications for its effectiveness in reducing suicidal ideation when compared with the treatment as usual.

Methods

Turkish migrants (first and second generation) with mild to moderate suicidal thoughts were recruited from the general population using social media and newspaper advertisements. The intervention group obtained direct access to a 6 weeks guided online intervention while participants in the waiting list condition had to wait for 6 weeks. The intervention is based on an existing online intervention and was culturally adapted. Participants in both conditions completed baseline, post-test, and follow up questionnaires on suicidal ideation (primary outcome), depression, worrying, hopelessness, suicide attempt and self-harm, acculturation, quality of life, and usability. In addition, participants were interviewed to examine the feasibility and mechanisms of action in more depth. The responses were analysed by inductive thematic analysis.

Results A total of 85 people signed up via the study website and 18 (10 intervention, 8 waitlist control) were included. While the therapeutic benefits were emphasised (e.g. feeling connected with the intervention), the feasibility was judged to be low. The main reasons given were: not having severe suicidal thoughts and not being represented by the culturally adapted intervention. No suicide attempts were recorded during the study. The suicidal ideation, depression and hopelessness scores were improved in both groups. Feasibility and Pilot RCT ― 199

Conclusion The online intervention, as might be expected in a small feasibility study, was not superior to the control condition. Future studies need to attend the implementation issues raised including measures of stigma, acculturation, careful cultural adaptations alongside more attention to coaching and relational support. They should also consider how to improve engagement alongside selection of those who are motivated to use online interventions and offer alternatives for those who are not.

7 200 ― Chapter 7

introdUction

uicide is a global public health problem with enormous consequences at individual and societal levels (Bertolote & Fleischmann, 2005). The international lifetime prevalence of suicidal ideation, plans and attempts in general population is 9.2%, 3.1% and 2.7%, respectively (Nock et al., 2008). SMembers of some ethnic groups are at higher risk for suicidal behaviours compared to others (Bhui et al., 2018). In Europe, Turkish populations are among the largest ethnic minority populations and they have disproportionate rates of suicidal behaviours compared to the ethnic majorities in their respective host countries (Bursztein Lipsicas et al., 2012). In the Netherlands, there is an elevated risk of suicidal ideation in Turkish adolescents (38.1%) when compared with ethnic Dutch (17.9%) adolescents (van Bergen et al., 2008). Suicide attempt risk is elevated 2-5 fold among Turkish migrant women aged between 14 and 25 when compared with same aged women from majority populations of the Netherlands (Burger et al., 2006), Germany (Lizardi et al., 2006) and Switzerland (Brückner et al., 2011). The causes of the increased prevalence rates are not clear yet. Some have argued that this might be associated with gender related factors such as domestic violence and “honour-related” violence which are important life events among women presenting suicidal behaviours worldwide (Heredia-Montesios et al., 2018; van Bergen et al., 2019). Others have stressed that people might encounter difficulties in their adaptation process to host countries, and interpersonal and structural discrimination within institutions, contributing to the elevated risk of suicidality (Schouler-Ocak, 2015; Bhui, 2010). Existing guidelines for treating suicidal behaviours in the UK and in the Netherlands recommend cognitive behavioural therapy (CBT) and CBT based interventions, such as Dialectical Behavioural Therapy (DBT), in managing suicidal behaviours which are delivered face-to-face (National Institute of Clinical Excellence, 2019; van Hemert et al., 2012). These interventions are based on a general cognitive model which is expanded to suicidal behaviours (Beck, 1995; Wenzel et al., 2009). How one interprets Feasibility and Pilot RCT ― 201 a situation determines emotional and behavioural reactions to those situations (Beck, 1995; Wenzel et al., 2009). Interpretation is distorted by errors in thinking such as overgeneralization and emotional reasoning. CBT interventions aim to restructure thinking errors, helping individuals to make sense of situations from a more realistic point of view (Beck, 1995). The efficacy of the Mentalization-Based Treatment (MBT) has also been demonstrated in several RCTs (Bateman & Fonagy, 1999; Rossouw & Fonagy, 2012) and longitudinal studies (Bateman & Fonagy, 2008) among adult and adolescent populations. Mentalization is the capacity to understand actions in terms of thoughts and feelings (Rossouw & Fonagy, 2012). When mentalizing is compromised in interpersonal relationships, negative thoughts are experienced in greater intensity leading to an urgent need for distraction. In this context, suicidal behaviours may serve as distraction (Rossouw & Fonagy, 2012). Not everyone receives these interventions and this is especially true for ethnic minorities (Bhui et al., 2018; Mcguire & Miranda, 2008). Cultural barriers such as stigma and shame attached to suicide might prevent them from utilising these interventions 7 (Eylem et al., 2016; Caplan, 2016). Additionally, poor language proficiency of the help- seeker and cultural mismatch between the mental health professional and the help-seeker often result in communication barriers during the help-seeking process (Morgan et al., 2009). It is therefore a global challenge to optimise these interventions for migrants and ethnic minorities (Rathod et al., 2015). In recent years, online interventions have been introduced into mental health services as an addition, or alternative for, face-to-face service delivery. These interventions can be delivered through PC’s, mobile phones or tablets and can be either guided or un- guided. Guidance can be delivered by a clinician or a trained coach (Cuijpers et al., 2017) and is aimed at motivating patients, to explain things which are unclear, and to provide feedback on the content of the sessions (Cuijpers et al., 2017). Unguided e-mental health interventions may involve automated feedback but do not provide any professional support related to the therapeutic content (Cuijpers et al., 2017). Guided interventions are often more effective than unguided interventions (Cuijpers et al., 2010) but unguided 202 ― Chapter 7 interventions are more scalable and can reach larger groups of people (Cuijpers et al., 2019). There is growing evidence supporting the value of online interventions in engaging with people from across the life span and from different ethnic backgrounds in treatment of common mental disorders such as depression and anxiety (Cuijpers et al., 2019). There are also several randomized controlled trials testing the effectiveness of online interventions for people with suicidal ideation (e.g. van Spijker et al., 2014; van Spijker et al., 2018; De Jaegere et al., 2019). Recently, meta-analytic studies indicated significant beneficial treatment effect of those interventions compared to treatment as usual (g = -0.26; 95% CI: -0.48, -0.03; Arshad et al., 2019). These studies however, have been carried out in general populations. Their effectiveness for even more vulnerable populations such as migrants and ethnic minorities with suicidal ideation is lacking (Mishara & Kerkhof, 2013; Shand et al., 2013; de Beurs et al., 2014; Caplan et al., 2018). In this study we have used an e-mental health intervention for suicidal ideation, which was developed for the general population in the Netherlands by van Spijker and colleagues (van Spijker et al., 2010). We have adapted this intervention for the Turkish migrants in the Netherlands and in the UK (Eylem et al., 2015; Eylem et al., 2016). Cultural adaptation is defined as ‘the systematic modification of an evidence-based treatment or intervention protocol to consider language, culture, and context in such a way that it is compatible with the individual’s cultural patterns, meanings, and values’ (Bernal, et al., 1995). CBT interventions offer flexibility to be adapted according to the needs and expectations of a diverse help-seeker population in multicultural health care (Rathod et al., 2015). Culturally adapted online interventions seem promising for ethnic minorities (Harper-Shehadeh et al., 2016) and for Turkish migrants specifically (Ünlü İnce et al., 2013, 2014c). The objectives of this pilot study are twofold: 1) to investigate the feasibility of the adapted online intervention among Turkish migrants in the UK and in the Netherlands; 2) and to investigate whether there is an indication of the effectiveness of the culturally Feasibility and Pilot RCT ― 203 adapted online intervention in reducing suicidal ideation when compared with treatment as usual. methods

Design, Settings and Participants This study is a randomised controlled trial in which patients were randomised to the guided online intervention or to a wait-list control group. The control group could directly access a website with psycho-education about the reasons for suicidal thoughts, risk factors and where to seek help (e.g. 113 online, Samaritans, mental health organisations) and could start with the intervention after the post-test measurement at 6 weeks. Among a sub-group of 12 participants who completed the intervention, semi- structured telephone interviews were held in order to obtain more in-depth information about their views on the intervention, the effects and ultimately on the feasibility of the intervention. Participants were recruited between January 2017 and October 2018. 7 Recruitment took place among the Turkish speaking population (i.e. people of Turkish, Kurdish and Turkish Cypriot background; Eylem et al., 2016; Ersanilli & Koopmans, 2010) in the Netherlands and in the UK. Participants were recruited from the general population through newspaper advertisements, TV programmes, banners on relevant websites (e.g. of community and health organisations), through social media and through public events. These events were organised in collaboration with the community organisations representing the population of interest. Eligible participants were18 years and older, and were Turkish (being born in Turkey or having at least one parent being born in Turkey). Everyone with a suicidal ideation score of 1 or higher on the Beck Scale for Suicide Ideation (BSS) scale was eligible for the study. Participants had to have access to a PC and internet. Finally, they needed to be willing to provide their name, telephone number and e-mail address of themselves as well as from their GP. These data were needed for our safety procedure. Already receiving help, regardless of the source, was not an exclusion criterion. 204 ― Chapter 7

A study website (http://kiymacanina.org/) was created where potential participants could find more information about the study. The information was presented in Turkish as well as in English and Dutch. Through this website patients could register. Those who registered received an e-mail with further information, an informed consent form and a link to the baseline questionnaire. Those who returned the informed consent, provided their contact details and those of their GP, filled out the baseline questionnaire, and did not fulfil our exclusion criteria were included. The randomisation scheme was derived using random allocation software by an independent researcher. Randomisation was stratified for the UK and the Netherlands and took place in a 1:1 ratio. The randomisation outcome was communicated to the participant by e-mail with either a log-in code for the intervention or a link to a website with general information on suicidality for the wait- list control group. The study protocol is described in more detail elsewhere (see Eylem et al., 2015) and was registered in the Netherlands Trial Register NTR5028 (see https:// www.trialregister.nl/trial/4926).

Ethics Statement This study was approved by the Medical Ethics Committee of the VU University Medical Centre in the Netherlands (registration number 2014. 187) and by the Queen Mary University of London Research Ethics Committee in the UK (registration number QMERC2014/46). Written informed consent of participants was obtained after the study and all procedures had been fully explained in writing. Participants could ask questions by e-mail or telephone if wanted.

Safety As this study involved vulnerable people who are at risk of suicide, a safety protocol was used (see Eylem et al., 2015 for further information). In summary: every participant (in both conditions) was asked to fill out the Beck Scale for Suicidal Ideation (BSS) once in two weeks throughout their participation. Whenever a participant exceeded the cut-off score of 29 on the BSS, we phoned the participant to perform a risk assessment. Feasibility and Pilot RCT ― 205

If necessary, we contacted the GP of the participant. The phone calls were made by a psychologist in the research team under supervision of a licensed clinical psychologist in the Netherlands and a licenced consultant psychiatrist in the UK who were both experienced in suicide prevention. Participants’ GPs were also contacted if a participant could not be reached.

Intervention The original version of the intervention is developed by van Spijker and colleagues (Kerkhof & van Spijker, 2011). RCTs into the unguided version of this intervention showed effectiveness of this intervention in reducing suicidal ideation compared to the treatment as usual in general Dutch (d=0.2 ; van Spijker et al., 2014) an Belgian populations (d= 0.34;De Jaegere et al., 2019), but it showed no effect when implemented in an Australian population (van Spijker, et al, 2018). The intervention is based on the CBT framework (Beck, 2005). Within this framework, some mindfulness exercises were also included. The main principle is that, 7 worry, rumination and repetitive suicidal ideation each produce obsessive attention to particular thoughts, sometimes resulting in a desire to end consciousness as a way to end the tantalizing repetition of suicidal thoughts (van Spijker, 2012). Thus, the aim of this intervention is to enhance controlled thinking (i.e. focusing on postponing worrisome thoughts to specific time-slots “worry times’’ of the day and not thinking of these thoughts for the rest of the day). The intervention consists of six modules, focusing on 1) the repetitive character of suicidal thoughts, 2) regulating intense emotions, 3) identifying automatic thoughts; 4) thinking patterns, 5) thought challenging, and 6) relapse prevention (see Eylem et al., 2015). Each module contains a theory section, a weekly assignment and several exercises. Weekly assignment is an essential element of the intervention encouraging participants to practice the techniques they learn in the intervention. For example, the first module explains that suicidal thoughts can develop out of self-protection, as keeping on living may seem worse than dying. Similarities between worry and suicidal thinking are also 206 ― Chapter 7 outlined. A weekly assignment involves tallying suicide-related thoughts to obtain an idea of how often these thoughts occur, while the exercises aim at learning to manage these repetitions better by introducing worry postponement. Participants are advised to do one module per week. Guidance was provided by coaches (2 in the UK and 2 in the Netherlands), who were supervised by a team of three experts including a psychologist, a licenced clinical psychologist (NL) and a consultant psychiatrist (UK). The coaches were bi-lingual (English-Turkish or Dutch-Turkish) and either students at a Masters level or practitioners seeing patients. They received training about how the online intervention works, safety protocol, referral system in both countries and their roles and responsibilities whilst providing guidance to the participants. Regular supervision meetings through skype and/or e-mail communications were arranged. Coaches provided online feedback to participants after their completion of each module. The aim of this personalised feedback was to help participants understand the exercises and homework assignments as explained in the lessons. Moreover, it was used to motivate the participants to continue with the intervention.

Cultural adaptation In the current study we adapted this intervention linguistically as well as culturally for the Turkish migrant populations. The decisions made during the cultural adaptation process are outlined in appendix B. In summary: we used the ecological validity model of Bernal et al (1995). This framework allows to retain the core principles of CBT in order to preserve treatment validity but also allows flexibility (Rathod et al., 2013). The model delineates eight dimensions when culturally adapting an intervention. These include the use of appropriate language, persons (cultural similarities/differences between the client and clinician which shape the therapeutic relationship), metaphors (symbols and concepts), content (cultural knowledge), concepts (treatment concepts that are culturally congruent), goals (that support adaptive cultural values), methods (cultural enhancement Feasibility and Pilot RCT ― 207 of treatment methods) and context (consideration of acculturation, social context) (see Appendix A). First, the intervention has been translated and back-translated. The consistency of the adaptations in three languages (English, Dutch and Turkish) has been checked by bi-lingual speakers. The cultural adaptation was based on the results of 6 focus groups and 7 individual interviews with 38 Turkish-speaking lay people and 4 professionals living in the Netherlands or in the UK during the year 2014/2015 (see Eylem et al., 2016). Based on this data we included some well-known idioms and metaphors describing psychological distress and suicide in Turkish language (see appendix B for examples). Further, we made some technical modifications in line with the preferences of the potential Turkish-speaking users such as reducing the text and providing more visual content (see appendix B for more examples). We also added cultural case examples in the sessions. As part of the adaptation, some theoretical modifications have also been made. In line with the well-documented evidence supporting the value of MBT (Rossouw & Fonagy, 2012) for instance, a well-known MBT based exercise called “safe place” was 7 incorporated into the crisis plan in the intervention. This exercise uses guided imaginary and encourages people to create an imaginary safe place that they could visit whenever they feel the need to be grounded (Rossouw & Fonagy, 2012) (see appendix A)

Deviations from the study protocol In the study protocol, we initially planned to do full RCT (see Eylem et al., 2015). The sample size of the RCT was calculated based on the expected effect of the primary outcome measure: the reduction in frequency and intensity of suicidal thoughts as measured with the BSS (d=0.40). The rationale for this decision was based on the trial of van Spijker who found an effect size of d=0.2 for an unguided version of the treatment (van Spijker et al., 2014). We expected a higher effect size since we provided personal coaching which generally leads to higher effect sizes (e.g. Cuijpers et al., 2010). Based on power of 0.80 and an alpha of 0.05, 100 participants were needed in each condition. 208 ― Chapter 7

Given the expected drop-out of 30%, the total sample size was determined as being 286 (Eylem et al., 2015). Although intended to be a definitive trial, it did not succeed and at best became a feasibility study with process evaluation to understand the components and how they operate. We conducted in depth interviews with the participants who completed the intervention and consented to be interviewed on feasibility issues. Furthermore, we have excluded the following questionnaires from the study: Suicidal Ideation Attribution Scale, the Hospital Anxiety and Depression Scale and an item measuring the satisfaction with the treatment. This was done to reduce the burden on participants and to increase their motivation to fill out our questionnaires. Moreover, we did not include three months follow-up assessments (T4) in the analyses and have done completers analyses. The reason was the underpowered study design which did not allow us to perform complex statistical procedures such as multiple imputation. We have used T4 to monitor the safety of the participants only. Finally, in the original protocol we described that we would add two additional modules on self-harm to the original intervention. In the end we decided to keep the focus on suicidal thoughts only (see Eylem et al., 2015).

Outcome Measures As part of the feasibility, the system usability scale was used and interviews were conducted. The feasibility was measured at post-test (6 weeks after baseline: T3). Usability of the online intervention was measured with the System Usability Scale (SUS). The SUS is composed of 10 statements that are scored on a 5-point scale of extent of agreement (score 0 to 100). The reliability is good (Cronbach’s alpha 0.91) (Bangor et al., 2008). Interventions with scores of 70 and above are accepted as highly usable (Bangor et al., 2008) and scores between 50 and 70 indicate acceptable usability of an intervention. Interventions with scores 50 and below are subject to concerns about their usability by the target population and should be investigated further (Bangor et al., 2008). Feasibility and Pilot RCT ― 209

In order to answer the research question about the indications of effectiveness we sent patients an online questionnaire. The primary outcome measure in this study is the reduction in the frequency and intensity of suicidal thoughts. This was measured with the Beck Scale for Suicidal Ideation, self-report (BSS) (Beck et al.,1988) at baseline (T0), 2 and 4 weeks into the intervention (T1 and T2), and at post-test (6 weeks after baseline: T3). The BSS is a 21-item measure assessing the severity of the suicidal ideation (Beck et al., 1974; Beck et al.,1979). Each item is scored from 0 to 2. The total score is obtained by adding the first 19 items and it ranges from 0 to 38. High score represents high suicidal ideation. The BSS has good psychometric properties in English (Beck et al., 1988, Beck et al., 1979) and Turkish (Dilbaz et al., 1995; Polatöz et al., 2011). Secondary outcome measures include self-report measures: depression, hopelessness, worrying, quality of life, self-harm behaviour and suicide attempt, acculturation, and user-friendliness and interviews measuring the acceptability and cultural relevance of the intervention. Depression was measured with the Beck Depression Inventory (BDI) comprised 7 of 21 items (Beck et al., 1961). Each item is scored from 0 to 3. The severity ranges from minimal depressed (score lower than 13) to severely depressed (scores between 29 and 63; Beck et al., 1961). It is a reliable and valid (Beck et al., 1961) measure for assessing depression. The BDI has been validated in Turkish and Dutch population (Hisli, 1987). Hopelessness was measured with the Beck Hopelessness Scale (BHS), which contains 20 true and false statements (Beck and Steer, 1988). Each statement is scored from 0 to 1 and the total score ranges from 0 to 20. A high score indicates high degree of hopelessness. The instrument has good psychometric properties (Brown, 2011). This is also true for the Turkish version (Seber et al., 1993). Worrying The Penn State Worry Questionnaire (PSWQ-PW) is a 15 item inventory assessing the weekly status of pathological worry (Stöbber and Bittencourt, 1998). Each item is scored on a 7 point rating scale, ranging from never 0 (never) to 6 (almost always). The total score ranges from 0 to 90 with a high score indicating more worrying. PSWQ- 210 ― Chapter 7

PW shows good reliability and convergent validity (Stöber and Bittencourt, 1998). The Turkish version demonstrated reliability (Yılmaz et al., 2008). Quality of life The Euro Quol (EQ-5D) is an instrument measuring health quality of life and comprised 5 items: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression (Rabin and Charro, 2001). Each item is required to be rated as 1 (no problem), 2 (some problem) or 3 (extreme problem). The current health state is also rated on a scale ranging from 0 (worst imaginable state) to 100 (best imaginable state). Both Dutch and Turkish versions have been validated (Lamers et al., 2005; Eser et al., 2007). Suicide Attempt and Self-Harm (SASH) Four questions measuring the previous suicide attempt and the presence of self-harm were taken from the original Self-Harm Questionnaire (Eylem, 2011). Acculturation was measured with an adapted version of the Lowlands Acculturation Scale (LAS). The original scale measures the degree of acculturation (Mooren, 2001) using 25 statements which represent the difficulties that migrants might face. On the basis of a six-point Likert scale, item scores range from 1 (not applicable) to 6 (very applicable). The instrument is validated among Turkish migrants living in the Netherlands (Mooren et al. 2001). It has been adapted to measure the 2 dimensions of acculturation: participation (4 items) and maintenance (11 items; Berry, 1997). The subscale participation measures tendency to participate in social life of the host country such as interacting with other minority and majority groups (score range 4 to 23). The subscale maintenance measures the tendency to maintain one`s culture of origin such as preferring to interact with people from same ethnic background (score range 11 to 60). Higher scores indicate a greater degree of participation and maintenance. The new sub scales were reliable (chronbach’s alpha 0.86 for both scales) and internally consistent (participation α = 0.79; maintenance α = 0.80; Eylem et al., 2019).

Analyses

We defined feasibility as engagement with the intervention (Arshad et al., 2019), its usability in daily life, as well as the potential for delivering a full trial in the future. Feasibility and Pilot RCT ― 211

The following components were assessed during the interviews in order to identify the facilitators and barriers influencing the engagement with the intervention. Cultural relevance (i.e. familiarity and relevance of the therapeutic content to one`s cultural background), cultural appropriateness (i.e. appropriateness of the therapeutic content in terms of the cultural context) and acceptance (i.e. feedback on the experience of using the intervention in real life (Cuijpers et al., 2019). The interviews were conducted with the participants who completed the intervention and consented to be interviewed (N=12) and started with an open question: ‘What was your overall experience during your participation?’ A topic guide was used for the remainder of the interview which was created on the basis of the relevant literature and discussions with the rest of the researchers taking part in the study. The interviews were approximately 30 minutes long and were recorded for verbatim transcription. We used thematic analysis as a means of identifying, analysing and reporting explanatory models, and for understanding which elements of the intervention facilitated or hindered participants` progress during their participation and how the intervention 7 can be optimised to increase its acceptability, relevance and user-friendliness (Braun & Clarke, 2006). Pseudonyms were assigned to each interviewee. First, the code system (and categories and themes developed on the basis of the coding process) was developed gradually and collaboratively. The code system was developed on theoretical grounds and included following categories: Definition of facilitators and barriers, specific examples for cultural relevance, appropriateness and acceptability of the intervention and specific recommendations for further improvement. Each of these categories had a number of sub-categories and codes. The definition of therapeutic gains as a result of using the intervention (e.g. therapeutic alliance) emerged during the analysis. This category was considered as important and decided to be analysed separately. The coding system was developed by the first author and was checked independently by a second person (Y.A.) who was not involved in the research. Once an agreement was reached, they were further developed, refined and applied to the transcripts. The first 212 ― Chapter 7 author was the main coder and Y.A. involved as a second coder, who systematically counter-checked the coding, to assure the robustness and the internal validity. The data was coded manually. Disagreements over the coding were discussed between the main and the second coder and where necessary experts (S.R.) were consulted. Detailed descriptive accounts were produced for each major theme alongside the related extracts from participants’ transcripts. Analysis continued until no new themes emerged from the transcripts (see table 4 for the themes). The RCT was carried out in accordance with the CONSORT guidelines (see consort checklist attached). First, t-tests and chi-square tests, as implemented in SPSS, were used to compare the baseline characteristics of those who were allocated in the intervention group with those who were allocated to the waitlist control group. Second, we tested for indications for effectiveness. We used Bayesian Repeated Measured ANOVA as implemented in the JASP (version 0.9.2), which is a free and open-source graphical programme for statistical analysis. The rationale for choosing Bayesian approach over the classical inferential approach is based on the following advantages: Bayesian approach considers all possible models (null model, between and within group differences) and assigns more weight to those models that predict data relatively well (Wagenmakers et al., 2018a; Wagenmakers et al., 2018b). Classical inferential model selects the best model, estimates its parameters and might produce overconfident conclusions on data by neglecting model uncertainty (Wagenmakers et al., 2018a; Wagenmakers et al., 2018b). Since the present study design is underpowered, we assumed that Bayesian approach would account for the uncertainty of the all possible models and would allow us to make more reliable conclusions based on the existing pilot data (Wagenmakers et al., 2018b). Additionally, we did a sensitivity analysis on those participants who reported severe suicidal ideation at baseline to see if the indications for an effect of the intervention were stronger than in the whole group including those with very mild suicidal ideation. Feasibility and Pilot RCT ― 213

resUlts

Participants Figure 1 shows the participant flow through the trial. A total of 85 people registered to the study website, while 50 people completed the baseline questionnaire. Of those 15 (30%) proved to be ineligible, mainly due to not having suicidal thoughts (N=6, 12%), not living either in the UK or in the Netherlands (N=6, 12%) or being younger than 18 at the time of the registration (N=3, 6%). The remaining 35 people were eligible but 17 of them (49%) failed to return their consent. The remaining 18 (51%) eligible respondents returned their consent forms and were randomised. We contacted all 18 participants for a telephone interview after ending the intervention. Two thirds (n=12; 66.7%) agreed.

7 214 ― Chapter 7

Figure 1. Participant flow through trial Feasibility and Pilot RCT ― 215

Table 1 displays baseline characteristics for all participants randomised. The majority was female (N=13; 72.2%) with paid employment (N=11, 61.1%). The participants were mostly single (N=11; 61.1%), mainly based in the UK (N=16; 88.8%) and half had a University degree (N=8; 44.4%). Mean age of the total sample was 33.5 years (SD=8.38). Half of the participants indicated not receiving any form of care at baseline (N=10; 55.5%), while some were seeing a psychologist (N=3, 16.6%), a GP (N=3, 16.6%), or a psychiatrist (N=1, 5.5%). A considerable number of participants were extremely dissatisfied with the previous psychological help (N=11, 61.1%). Among those who were receiving help, only one of them (5.5%) evaluated it as helpful. There was a significant difference between intervention and waitlist control group participants on the level of acculturation. Participants in the waitlist control group more often maintained their own ethnic culture compared to the participants in the intervention group (p < 0.01) who had more often participated in the social life in the Netherlands and/or the UK. On average participants experienced mild levels of suicidal thoughts (M=13.50, SD=8.07). Out of 18 participants, 7 scored 20 and higher on the BSS, indicating severe 7 suicidal thoughts (38.9%; van Spijker et al., 2014), and 11 of them (61.1%) scored below 20 on the BSS indicating mild suicidal thoughts. On average there were substantial depression symptoms (M=29.61, SD=10.16), feelings of hopelessness (M=12.94, SD=4.39) and worry (M=67.27, SD=17.61). There were no statistically significant differences at baseline between the two conditions regarding these clinical characteristics. 216 ― Chapter 7

Table 1. Baseline characteristics Intervention Waitlist-Control Participants Participants Total P Characteristics (N=10) (N=8) (N=18) Value* Age (M,SD) 34.70 (3.81) 32.00 (9.35) 33.50 (8.38) 0.51 Gender, male (N,%) 3 (30%) 2 (25%) 5 (27.7%) 0.81 Type of Recruitment (N,%) 0.47 Facebook 8 (80%) 5 (62.5%) 13 (72.2%) Through a friend 2 (20%) 2 (25%) 4 (22.2%) Through a newspaper add 0 (0%) 1 (12.5%) 1 (5.5%) Employment Status (N,%) 0.96 In a paid employment 3 (30%) 2 (25%) 5 (27.7%) In an unpaid internship 1 (10%) 1 (12.5%) 2 (11.1%) Unemployed 6 (60%) 5 (62.5%) 11 (61.1%) Relationship Status (N,%) 0.65 Single 7 (70%) 4 (50%) 11 (61.1%) In a relationship 0 (0%) 1 (12.5%) 1 (5.5%) Married 2 (20%) 2 (25%) 4 (22.2%) Widow 1 (10%) 1 (12.5%) 2 (11.1%) Education Level (N,%) 0.86 Secondary school 2 (20%) 2 (25%) 4 (22.2%) University 5 (50%) 3 (37.5%) 8 (44.4%) Other 3 (30%) 3 (37.5%) 6 (33.3%) Help-seeking Status (N,%) 0.39 No help 7 (70%) 3 (37.5%) 10 (55.5%) GP 1 (10%) 2 (25%) 3 (16.6%) Psychologist 2 (20%) 1 (12.5%) 3 (16.6%) Psychiatrist 0 (0%) 1 (12.5%) 1 (5.5%) Other 0 (0%) 1 (12.5%) 1 (5.5%) Satisfaction with previous help (N,%) 0.15 Extremely unhelpful 8 (80%) 3 (37.5%) 11 (61.1%) Neither helpful /unhelpful 2 (20%) 4 (50%) 6 (33.3%) Helpful 0 (0%) 1 (12.5%) 1 (5.5%) Acculturation (M,SD) Participation 13.00 (4.57) 15.50 (3.42) 14.11 (4.18) 0.21 Maintenance 35.20 (6.26) 43.87 (4.05) 39.05 (6.87) 0.004* Feasibility and Pilot RCT ― 217

Table 1. Continued. Intervention Waitlist-Control Participants Participants Total P Characteristics (N=10) (N=8) (N=18) Value* Suicidal Ideation (M,SD) 12.30 (8.48) 14.87 (8.07) 13.50 (8.07) 0.52 Suicide Attempt and Self-Harm (M,SD) 2.10 (1.91) 1.50 (1.69) 1.83 (1.79) 0.50 Depression (M,SD) 26.80 (9.73) 33.17 (10.17) 29.61 (10.16) 0.19 Hopelessness (M,SD) 12.50 (4.08) 13.50 (4.98) 12.94 (4.39) 0.64 Worry (M,SD) 63.50 (19.92) 72.00 (4.05) 67.27 (17.61) 0.32 Quality of Life (M,SD) 8.30 (1.63) 7.62 (1.40) 8.00 (1.53) 0.36 *P values are based on t-test or Pearson X2 test

Safety We did monitor all the participants carefully through guidance and assessments (t1 and t2: bi-weekly assessments; t3: post-test; t4: 3 months follow-up). Participants indicated mild levels of suicidal thoughts at baseline (M=13.50; SD=8.07). We did not call any of 7 the participants as none of the participants exceeded the cut-off score of 29 on the BSS at any time during their participation. Thus, the safety protocol was never activated. There were no suicide attempts or suicides during their participation in the study.

Table 2. Mean Changes from Baseline to Post-test and Follow-up (N=16) Intervention (n=10) Control (n=6) Post-test Post-test Outcomes Pre-test (6 weeks) Pre-test (6 weeks) BSS (suicidal ideation; M, SD) 12.30 (8.48) 6.10 (4.50) 12.33 (7.60) 6.83 (3.86) BDI (depression; M,SD) 26.80 (9.73) 16.10 (6.96) 32.50 (10.84) 25.10 (8.67) PSWQ (worrying; M,SD) 63.50 (19.92) 58.40 (17.24) 73.33 (15.34) 58.17 (15.52) SASH (suicide attempt & self-harm) (M,SD) 2.10 (1.91) 0.90 (1.59) 0.83 (1.32) 0.83 (1.16) BHI (Hopelessness; M,SD) 12.50 (4.08) 7.80 (3.67) 12.66 (5.46) 10.83 (4.26) EQ5SD (Quality of Life; M,SD) 8.30 (1.63) 7.90 (2.28) 7.33 (1.21) 6.83 (1.16) 218 ― Chapter 7

feasibility

Quantitative Findings

Acceptability Overall, 8 out of 10 intervention participants (80 %) and 3 out of 8 participants (37.5%) in the waitlist control group completed all the sessions (see figure 1). Out of 6 sessions, the average number of completed sessions in the intervention group was M=5.6, (SD=0.9) while this was M=3.4 (SD=2.3) for the wait-list control group.

Usefulness Participants in the intervention group reported an average score of M=36.20 (SD=5.84) on the System Usability Scale. Participants in the waitlist control group scored M=29.16 on average (SD=6.52). Both scores are below the cut-off of what might be considered a useful intervention.

Qualitative Findings The thematic analysis on the 12 interviews identified 3 overarching themes (see table 3).

Table 3. Themes related to views on the internet intervention (Kıyma Canına) and its adapted content N=12 (9 women, 3 men) of Turkish descent, and aged 23-56, in the Netherlands and in the UK Themes N=12 N (%) 1. Therapeutic change 12 (100) Therapeutic alliance 9 (75) Self-management 10 (83) 2. The gap between reading it and doing it in real life 7 (58) Feeling connected 7 (58) Not feeling connected 5 (42) 3. Recommendations for improvement 12 (100) More diversity 9 (75) More directive approach 8 (67) Feasibility and Pilot RCT ― 219

Theme 1: Therapeutic change

Therapeutic Alliance All participants commented on developing therapeutic alliance (i.e. the relationship between coaches and the participants) as a result of using the intervention. Many participants identified personalised feedback as one of the key components helping them to benefit from the intervention. For some participants, personalised feedback did not only motivate them to continue but it also provided a safe environment to disclose their experiences:

“Receiving feedback was like exchanging letters with someone….Sometimes you cannot talk to

everyone about certain things. But receiving feedback and being able to respond to it, was like a

relief …. As I went through them, I kept on discovering new things about myself” participant A.

Self-Management Several participants were going through important life events for instance, domestic 7 violence, loss of a loved one and work related stress which resulted in crises. Crisis was often defined as feeling confused about how to handle stress. The psycho-educational aspect of the exercises and the feedback helped them to understand these crisis situations:

“I was going through a trauma….and was not able to make much sense of what was happening to

me…the feedback helped me to make sense of it all. It helped me to explain things from a scientific

point of view” participant D

Almost all participants emphasised better self-management as one of the most important benefits of the intervention. Several participants mentioned that the exercises about worry time were helpful in terms of managing the crisis situations. More specifically, worry time helped them to feel more in control of their thoughts and this was felt as an 220 ― Chapter 7 important accomplishment. Many participants emphasised this feeling as an important source of motivation:

“Sometimes when I was panicking about something, I was letting myself to worry…to think of

the worst case scenario that could happen to me… I was thinking of that for 10 minutes or so and

afterwards I was able to feel better…..Managing to do that was really helpful…..” participant F

“ It [following the intervention] gave me some peace of mind as I was doing something at least”

Participant H

Theme 2: The gap between “reading it” and “doing it” in real life

Feeling connected Feeling connected with the intervention and/or with the personal coach appeared as a strong facilitator. Being able to relate to the content helped them feel connected with the intervention. Suitability of the intervention was emphasised as a strong facilitator for feeling connected:

“I felt the exercises were suitable with my life style…Working with a coach was also helpful in terms

of feeling connected… I feel I gained skills that I could use for the rest of my life” Participant A

Some participants named mindfulness and mentalisation exercises as the most helpful ones in terms of their recovery. Most commonly mentioned reasoning was that such exercises were not restricted with the context and were easy to follow during the day:

“Mindfulness exercises [imagining your thoughts as if they were clouds and watching them pass

by] are not restricted with the context…so you can do them when you are sitting at the office or

when you are doing yoga…..’’ Participant D Feasibility and Pilot RCT ― 221

All participants spoke about feeling familiar with the culturally adapted content such as cultural case examples and the well-known metaphors explaining psychological distress and crisis situations. Several participants identified cultural familiarity as a pleasant experience helping them to feel more connected with the intervention. Those who feel connected were also able to relate to the intervention (i.e. cultural relevance) and often found it appropriate (i.e. culturally appropriate):

“The intervention was very familiar and it felt like I was not only getting professional help but was

also talking to a friend” Participant E

“I think all the examples were appropriate to the Turkish culture…They were also representative of

the types of problems that migrant populations are likely to face” Participant G

In terms of the acceptability of the intervention, participants expressed contradictory opinions. For some of them, the “self-help” principles made it difficult to use the 7 intervention. They emphasised that guidance and cultural adaptations were not sufficient:

“The difficulty with the online therapy is that, we need to do things on our own. When you see

a psychologist….when there is a person in front of you….you feel more in control….I think there

should be a psychologist in front of you and you should feel pushed….Do you see my point? After all

not feeling in control is the main reason why we need psychological help….Isn`t it?” Participant H

Not feeling connected Those who did not consider themselves as having severe suicidal ideation, did not feel connected with the intervention. They often reported feeling uncomfortable when they thought they were being considered as a severe case: 222 ― Chapter 7

“ The intervention was for severe cases [people who have intense thoughts about suicide]….I am not

in that group….. so sometimes the questions and the exercises were not so relevant to me. I asked

myself if this is how they really think about me. Am I considered as a ‘’nut case’’? This was affecting

my willingness to participate….You know…how you feel changes your decisions….” Participant H

Not feeling connected was also mentioned in relation to the culturally adapted content. Some participants mentioned that the adaptations such as cultural case examples, were representative of a group of traditional people and they were not able to relate to this specific group. They were more ambivalent especially about the appropriateness of the cultural case examples (i.e. adaptations in the context of the intervention):

“I don`t know…..Someone`s daughter broke up with her fiancé and so and so forth [cultural case

example]…I only laugh at such things when I hear them…they do not fit with my philosophy…”

Participant B

“The cultural case examples attracted my attention….Because my life experiences are different, I felt

awkward sometimes” Participant C

Theme 3: Recommendations for improvement

More diversity Many participants recommended to include more diversity in the context (e.g. more case examples representing different backgrounds). All participants commented on the usefulness of the online diary, which was part of the crisis plan encouraging participants to monitor their suicidal thoughts on a daily basis. For many participants online diary was frustrating as it was not representative of the variety of feelings they experience throughout the day and was also not practical:

‘’I struggled with the online diary…. I was asked to upload a picture representing my typical mood

of the day….But finding a specific picture representing a specific feeling was not really feasible for Feasibility and Pilot RCT ― 223

me…..The options for feelings were also too generic. I feel variety of emotions during the day not

only sad, angry or happy…..I felt I was not able to express myself there” Participant C

More directive approach Several participants spoke about including more instructions and coaching in the sessions. This was often mentioned as a way of feeling more in control whilst implementing the intervention in their daily life:

“There were many exercises…..and I needed to find out which one works better for me…. I didn`t

quite catch that in the beginning…It worried me… I felt I was not in control….I think there could

be more personalised guidance so that, it`s easier to find the right exercises” Participant I indications of effectiveness Overall, the analyses showed indication for change in suicidal ideation, hopelessness and depression scores, but not for worrying and quality of life scores over time (see table 2). 7 These improvements occurred in the intervention group as well as in the control group and there were no post-test differences between the two groups (Figure 2). A Bayesian two-way Repeated Measures ANOVA comparing the pre (T0) and post- test (T3) BSS scores of intervention and wait-list control group participants revealed indication for a time effect on suicidal ideation (BF10: 50.4), hopelessness (BF10: 14.44) and depression (BF10: 127.09) scores, but not for group or interaction effects (see table 3). The interaction plot for the suicidal ideation scores (see figure 2) indicates slightly more improvement in suicidal ideation among those in the intervention group compared to those in the wait-list control group. However, this was not supported by the results of the Bayesian analyses. We have repeated the Bayesian two-way Repeated Measures ANOVA on those who scored 20 and above on BSS scores only (N=7). The results indicated stronger indication for a time effect on suicidal ideation, and depression scores, but not for group or interaction effects (see table 5 in appendix B and interaction plots in appendix C). There 224 ― Chapter 7 were two exceptions. Among those with severe suicidal ideation, there was stronger indication for time and group effects on hopelessness scores (BF10:10.512). This means that it is strongly likely that among those with severe suicidal ideation in both groups, the hopelessness scores were improved over time. There was also strong indication for an interaction effect on quality of life (BF10:8.176). Among those with severe suicidal ideation in the intervention group, there is strong indication for a greater improvement in quality of life compared to those with severe suicidal ideation scores in the waitlist control group.

Figure 2. The interaction plot between group status and time (pre-test, week 2, week 4 and post-test) indicating changes in suicidal thinking in intervention group when compared with the wait-list control group Feasibility and Pilot RCT ― 225

Table 4. Model Comparisons Between the Null and the Alternative Models for the Study Variables (N=16) Variable Model BF10 Ratio BSS (T0 vs T3) Null model 1.00 a Time 50.41 1.00 Group 0.46 0.01 Time + Group 27.27 0.54 Time+Group+Time*Group 11.89 0.23 BSS (T1 vs T2) Null Model 1.00 a Time 1.32 1.00b Group 0.60 0.45 Time+Group 0.81 0.61 Time+Group+Time*Group 0.46 0.35 BDI (T0 vs T3) Null Model 1.00 a Time 99.84 1.00b Group 1.06 0.01 Time+Group 127.09 1.27c Time+Group+Time*Group 70.34 0.70 BHI (T0 vs T3) Null Model 1.00 a Time 14.44 1.00b Group 0.54 0.04 Time+Group 8.30 0.57 Time+Group+Time*Group 7.27 0.50 PSWQ (T0 vs T3) Null Model 1.00 a 7 Time 2.85 1.00b Group 0.57 0.20 Time+Group 1.72 0.60 Time+Group+Time*Group 1.29 0.45 EQ5SD (T0 vs T3) Null Model 1.00 a Time 0.43 1.00b Group 0.77 1.79c Time+Group 0.33 0.76 Time+Group+Time*Group 0.14 0.32 Note. T0: Baseline (pre-test); T1: Bi-weekly measures of BSS at week 2; T2: Bi-weekly measures of BSS at week 4; T3: post-test; BF10: Bayesian factor grading the intensity of the evidence supporting the alternative model against the null model Ratio: This column represents the ratio (the likelihood) of the effect of time against the group, time and group and the interaction models. The time model is denominator. The BF10 of each model has been divided by the BF10 of the time model in order to calculate the ratio of each model when compared with the time model. a: The ratio for the null model was irrelevant b:The ratio of time against time is always 1 c: When the ratio is between 0 and 1, there is a weak evidence supporting the alternative model against the time model. When it is greater than 1, that means there is a stronger evidence supporting the alternative model against the time model. 226 ― Chapter 7

discUssion This small randomized controlled trial was set out to investigate the feasibility, and to demonstrate indications of effect, of an e-mental health intervention for suicidal ideation which is culturally adapted according to the Turkish migrants in the Netherlands and in the UK. We included 18 participants and among those, 7 of them had substantial levels of suicidal thoughts at baseline. Despite this, no one reported a risk for a suicide attempt and the safety protocol did not need to be activated. Although therapeutic benefits of the intervention were emphasised (e.g. feeling connected with the intervention and/or personal coach), the low scores on usability of the intervention showed a number of barriers compromising its usability in daily life. Those with mild suicidal ideation and those who were not relating to the culturally adapted content (e.g. specific cultural case examples) emphasised not being represented by the content of the intervention. Further, there was no indication that the intervention group lead to better health outcomes than the control group. In both groups there was a reduction in suicidal ideation, depression and hopelessness, but not in worrying and quality of life. Participants in the study The reasons for the low up-take of the intervention remain unclear. One possible explanation is the difficulty in engaging with the target population during the recruitment process. In line with the trial of Ünlü İnce and colleagues (Ünlü et al., 2014b), various channels have been used during the recruitment process such as social and mainstream media promotion, and TV programmes. Additionally, public events have been organised such as exhibition and film discussion. The rationale for using the latter methods was the growing evidence supporting the value of them in engaging with the target group concerning topics related with stigma and shame (Belone et al., 2016). Even though such events were organised in collaboration with community organisations and were usually well-attended, they did increase the publicity but did not often result in more participants. Lack of anonymity during the recruitment process might be another important barrier. As part of the safety protocol, we collected personal information (name, address) Feasibility and Pilot RCT ― 227 and GP details during the recruitment process. Qualitative evidence on cultural meaning of suicide among Turkish migrants indicate that disclosing suicidal thoughts might mean dishonouring family and community by failing their expectations (Heredia-Montesinos et al., 2018; Eylem et al., 2016). Thus, the fear of disclosing identity might not have been eliminated during the recruitment process. Paper consent procedures in this current study might have contributed to this fear. Further, it might also be that our target population was reserved in terms of seeking help for suicidal thoughts. Although the risk for suicidal behaviours is well-documented among Turkish populations (van Bergen et al., 2019), their mental health literacy (i.e. recognizing suicidal thoughts and symptoms of psychological distress) might be low. Support for the deleterious impact of the low mental health literacy on participant engagement comes from a recent pilot study investigating the feasibility of a mobile app in treating depression among Hispanic population in the States (Caplan et al., 2018). Low mental health literacy was identified as a barrier restricting the up-take of the online app in this community (Caplan et al., 2018). Although appealing, the mental health 7 literacy was not investigated in our study. Thus, its impact in the implementation of our intervention remains unclear. There might be other human factors contributing to the low up-take of the intervention. For instance, the target population might not have wanted an online intervention. The negative general perception of online interventions (e.g. they might be more appropriate to people from higher social class or those who are younger or more educated) was an important barrier hindering the engagement with the target population in a pilot RCT in Lebanon (Harper Shehadeh et al., 2016). Another attitudinal barrier might be about participating in an experimental study (with lots of extra questionnaires and the chance of entering the control group). Participating in research might precipitate the fear that their personal information might be used against them by the institutions of the host countries such as not hiring them for a job position because of having suicidal thinking or symptoms of depression (Jansen-Kallenberg et al., 2017). 228 ― Chapter 7

Feasibility Feasibility issues point to the barriers during the implementation of the intervention in real life. One of the main barriers was not feeling represented, despite the cultural adaptations of the intervention content. Tseng has argued for ensuring philosophical adaptations are also considered, that is the meaning of therapy and a therapeutic relationship, alongside technical, theoretical and practical modifications (Tseng & Streltzer, 2004). More specifically, the characters representing Turkish migrants in the case examples were defined as more traditional in adherence to cultural and religious practices than the participants, and less educated than some participants. It could be that some of the participants in our sample had different cultural values and norms, emphasising caution in cultural adaptations to suit a range of demographic influences and cultural transition or acculturation. The main reason for contextualizing the original intervention according to the Turkish cultural norms and values was to reach out to those who are more attached to these values (i.e. more traditional) in particular. Research from the Netherlands (Eylem et al., 2016), Belgium (Güngör, 2008) and Germany (Aichberger et al., 2015) emphasise that Turkish migrants who are more traditional, are more at risk for suicidal behaviours and are less likely to access the available treatment. It could be that the online form of delivery of the intervention was not suitable for this group specifically, or that the groups in the Netherlands are more traditional than the majority of participants who were from the UK. This necessitates a measure of acculturation in future research studies and adaptation to suit a variety of cultural groups. There is a preliminary evidence for the acceptability of a culturally adapted group intervention targeting first generation Turkish migrants in the UK (Perry et al., 2019). Similar to the content of our online intervention, Perry and colleagues adapted the context of their intervention by incorporating idioms of distress and cultural case examples (Perry et al., 2019). Their intervention was delivered in a group format in a community setting which was accessible especially for the first generation migrants (Perry et al., 2019). Since the majority of participants considered the adaptations Feasibility and Pilot RCT ― 229 appropriate to the therapeutic needs of the traditional group of participants, it could be that not the content of adaptations but the format of delivery restricted the engagement with this group of participants. It could also be that in our study, the cultural adaptations on the context of the intervention underscored the diversity of the Turkish migrant populations in Europe. This finding points to the “overgeneralisation” as one of the main challenges encountered in the cultural adaptation literature (Rathod et al., 2015). Theoretically, it could be argued that even though migrants might share the same cultural identity (e.g. ethnicity, nationality and social group), there might be differences in micro-identities (e.g. political views, religion, sexual orientation) embedded within their cultural identities (Bhugra, 2005). The settlement and acculturation processes might complicate the formation of these identities even more so when the transnational migrant populations, such as Turkish migrants, are concerned. For instance, there has been an increase in political refugees from Turkey in West Europe (e.g. Germany, the Netherlands, and the UK) in recent years (Özdemir et al., 2019). This new group of Turkish migrants might have 7 different micro-identities compared to the Turkish migrant populations with a long- standing history of settlement who have strong ties with their heritage culture (Özdemir et al., 2019). Even though we do not have information about the migration history of the participants, it might be that our sample was more representative of the recent group and this might have restricted the usability of the intervention in their daily life. Another reason for feeling not represented was identified as not having severe suicidal thoughts. It could be that our intervention was more suitable to those with more severe suicidal ideation. Support for this comes from the results of our sensitivity analyses. We found stronger indications for an improvement in suicidal ideation, depression and hopelessness scores in both groups when we restricted the analyses with those with severe suicidal ideation scores. We also found stronger indication for a greater improvement in quality of life among those with severe suicidal ideation in the intervention group compared to those with severe suicidal ideation in the waitlist control group. These findings suggest that our inclusion criteria (scoring 1 and above on BSS) was too broad. 230 ― Chapter 7

This is discrepant from the trial of van Spijker and colleagues (van Spijker et al., 2018) where severe levels of suicidality and depression were associated with poor motivation (van Spijker et al., 2018). These differences could be explained with the cultural differences in help-seeking patterns. For instance, it is well-known that Turkish migrants tend to delay help-seeking until the symptoms of distress are more severe (Ünlü İnce et al., 2014b; Fassaert et al., 2009). It could be that when the psychological distress caused by suicidal thoughts is unbearable then stigma and shame attached to suicide might have less influence on their motivation to seek help. However, this remains inconclusive for a number of reasons. First of all, we did not measure the mental illness stigma among participants. We also did not investigate whether there was a relationship between intensity of suicidal ideation, stigma and help-seeking for suicide among participants. Additionally, we did not have a sufficient number of participants to consider this trial definitive and assess effectiveness. Feasibility issues might also be related with the delivery of the intervention. Some participants recommended a more directive approach giving them more clear instructions in following the sessions. Our intervention was set out with a lot of different exercises and participants were encouraged to choose the appropriate ones for them. It might be argued that more choice is not always the best option especially for some ethnic groups who are seeking explicit advice and assertion in their help-seeking process (Knaevelsrud et al., 2015). Effects One of the obvious reasons for not finding an indication for the effect of the intervention was the small sample size. However, since the scores of the two groups were almost identical we do not expect that a larger trial will reveal effects. Both groups improved over time. The finding that there was an improvement in suicidal ideation scores and in depression in the control group is in line with the results of the other trials (van Spijker et al., 2014; van Spijker et al., 2018; De Jaegere et al., 2019). One of the explanations could be the effect of the safety protocol itself (i.e. being measured bi-weekly, having a safety Feasibility and Pilot RCT ― 231 procedure installed when necessary) in the waitlist control group during the first 6 weeks period upon randomization. It could be that safety protocol functions as an intervention in itself in the waitlist control group, and might be separately investigated. There is growing evidence indicating the value of safety planning as a standalone psycho- educational intervention increasing awareness about crisis situations, warning signs and available services for further help (Stanley and Brown, 2012). Alternatively, improvements in both groups can be attributed to receiving usual care which cannot be ruled out in the present study. Even though the majority of participants in both groups indicated not receiving any other help at baseline, their exposure to other sources of information and/or help was not carefully monitored throughout the study. This is an important methodological limitation (van Spijker et al., 2018; De Jaegere et al., 2019; Caplan et al., 2018). Future studies could benefit from an attention-control condition to monitor the elements and effects of usual care (De Jaegere et al., 2019). Another possibility might just be the passing of time in that many states of distress are self-limiting and people might just spontaneously recover (Cuijpers and Christea, 2015). 7 Notwithstanding with these limitations, this study has various strengths. An important strength is the safety of our intervention. The safety protocol did not need to be activated because of a risk for a suicide attempt during the study. None of the participants reported any adverse effects of the intervention, such as increased intensity of suicidal thinking, as a result of their participation. The safety of the e-mental health intervention is in line with the other trials investigating its effectiveness in general populations in the Netherlands (van Spijker et al., 2014), in Belgium (De Jaegere et al., 2019) and in Australia (van Spijker et al., 2018). Thus, vulnerable groups such as minority groups with severe mental health symptoms could be studied safely, but a sub- group needs to be better identified for maximal benefit. Importantly, others may prefer alternatives, so e-interventions cannot address the needs of all. Another strength is that the study followed a particular theoretical framework and a systematic approach to adapt the intervention according to the cultural beliefs and attributions about suicide and help-seeking for suicide among this high-risk group. 232 ― Chapter 7

The challenges encountered during the adaptation process such as overgeneralisation, highlight the importance of investigating the components of effective cultural adaptation further. For instance, when the context of the intervention is adapted for a particular sub-group within the target population, the fidelity of the intervention might be compromised and thus it might not be effective. It is also noteworthy that the majority of the participants emphasised the therapeutic effect of guidance. Guidance was often considered as helpful in terms of feeling connected with the intervention. One of the notable recommendations of participants was about incorporating more guidance in the intervention. The expectation for more directive approach is in line with the cultural adaptation practices for Muslim populations (Knaevelsrud et al., 2015). The refusal to give explicit advice or lacking assertion has been associated with incompetence and indecisiveness of the mental health professional, which might lead to the patient becoming irritated or discontinuing their therapy (Knaevelsrud et al., 2015). This finding supports the rationale of the decision for incorporating guidance into the e-mental health intervention in the present study as being different to the other trials (van Spijker et al., 2014; De Jaegere et al., 2019; Mühlmann et al., 2017).

Implications and Conclusions: To summarise, the present study is an important step to further the current knowledge on whether online interventions could provide a feasible and an effective alternative in more complex contexts, including ethnic minority groups who are at elevated risk for suicidal behaviours. Our results suggest that there are number of considerations to incorporate into the intervention itself and into the design before further effectiveness testing. One of the key lessons for future studies is to investigate which components of cultural adaptation are effective in improving engagement with the target group. Our results suggest that adaptations in the context of the intervention should not exclusively represent the norms and values of the settled community but also the values of those who are in the process of settlement. Another key lesson for optimising the intervention further for the Turkish migrant populations is to structure the content of the intervention Feasibility and Pilot RCT ― 233 more. This might be important especially for ethnic groups who seek for instructions in their help-seeking process. Since the intervention was safe to use in our study, as well as in other trials, the intervention can be implemented through anonymous online platforms such as Samaritans in the UK and 113 online in the Netherlands. This might remove barriers for instance, fear of disclosing identity and fear of dishonouring one`s family, which were highlighted as potential barriers hindering participation in the present study. Since the participants valued the process of guidance, the online intervention might be well suited as an add – on intervention to regular psychotherapeutic treatment. The effectiveness of such an additional intervention to regular face to face treatment should be studied carefully. To conclude, the challenges encountered in this pilot study can be viewed as part of the incremental steps necessary to build future success in implementing online interventions among ethnic minorities in treatment of suicidal behaviours. This study and the previous trials seem to suggest the intervention needs further work and refinement and there should be more feasibility and exploratory trials of a modified intervention, refine it 7 iteratively with feedback.

8

general discUssion 236 ― Chapter 8

introdUction

n this final chapter, the key findings are summarised, study limitations are described and the results are compared with prior research findings. Additionally, implications for clinical practice are discussed, and future directions for research I are highlighted. key findings and previoUs research

Ethnicity and mental illness stigma There is growing evidence highlighting the detrimental consequences of mental illness stigma for racial and/or ethnic minorities relative to majorities (Nadeem et al., 2007; Bhui, 2010; Roberts et al., 2018). However, none of the existing reviews in the stigma literature focus on the racial and/or ethnic differences in mental illness stigma. This limits current understandings of the impact of stigma for culturally diverse populations. Thus, we conducted a systematic review and meta-analysis (Chapter 2) to investigate the racial differences in mental illness stigma. We also investigated if certain study characteristics: racial classification, quality of the studies and types of stigma outcomes (self-report vs vignette) were related with the mental illness stigma. The results showed significantly more mental illness stigma for CMDs among racial minorities, regardless of the race, compared to majorities (g=0.20). Additionally, the number of studies with high risk of bias had higher effect sizes than the studies with lower risk of bias. These findings are in line with previous research indicating more mental illness stigma among racial and /or ethnic minorities compared to majorities (e.g. Miranda et al., 2015; Subramaniam et al., 2017; Nadeem et al., 2007). In light of the intersectional impact of mental illness stigma (Bhui 2019), we argue that social adversities such as low socio-economic background, and belonging to more than one stigmatising groups such as being an ethnic minority and having a mental illness, might have collective impact on psychological distress (Bhui, 2019). Additionally, when stigma coincides with cultural General discussion ― 237 attributions discouraging help-seeking for mental health problems, individuals might be more likely to experience harmful effects of stigma such as unmet mental health care needs (Augsberger et al., 2015; Nadeem et al., 2007; Bhui, 2019). The finding that racial classification was not related with mental illness stigma differed from our expectations. This points to the complexity of the relationship between stigma and ethnicity (Babtiste, 2017). For instance, possible differences in experience of mental illness stigma between native born minorities, migrants and non-minorities were indicated (e.g. Makowski & von dem Knesebeck 2017). Thus, it could be argued that there are other important factors influencing the impact of mental illness stigma such as degree of acculturation in a host country or presence of a mental illness diagnosis, among some ethnic groups relative to the others (Bhui, 2019). At present, these are overlooked in the present stigma literature, and are not investigated in Chapter 2. Arguably, even though the content of the different types of stigma are different, they do happen in the same socio-cultural context and might have a similar mechanism (Fox et al., 2018). One of the implications of the meta-analysis described in Chapter 2 could be that it is crucial to understand the socio-cultural context such as the degree of acculturation and the cultural conceptualisation of mental illnesses predisposing some 8 ethnic groups to an increased risk for mental health problems and suicide relative to the others. Thus, in Chapters 3, 4 and 5 these factors: acculturation, cultural meaning attributed to suicide and the relevance of the cultural continuity of risk factors and characteristics of suicidal behaviours from the country of origin, were investigated in order to further the understanding of suicidality of Turish migrants.

Acculturation There is well-documented epidemiological evidence indicating the significance of acculturation difficulties on the mental health of Turkish migrants in a number of countries in Europe (Germany, Belgium, Netherlands; Aichberger et al., 2015; Phalet & Schonpflug, 2001; Fassaert et al., 2010). However, little is known about how specific acculturation strategies generate an increased risk of suicidal behaviours among Turkish 238 ― Chapter 8 migrants and in what circumstances this risk is increased and/or decreased. To address this knowledge gap in the literature, Chapter 3 investigated the relationship between acculturation and suicidal ideation among Turkish migrants in the Netherlands. In this chapter, the focus is placed on expanding the current knowledge base of theoretical models by examining the possible mediating role of hopelessness and the moderating role of attachment. The results showed that higher participation in the social life of the host country was associated with less hopelessness and less suicidal ideation. Greater maintenance of one’s ethnic culture was associated with higher hopelessness and higher suicidal ideation. Greater participation in the host culture was associated with less suicidal ideation particularly among those with less secure attachment styles. The above mentioned results are in line with previous cross-sectional research among Turkish migrants in Belgium (Phalet & Schonpflug, 2001), in Germany (Aichberger et al., 2015) and in the Netherlands (Fassaert et al., 2010) highlighting that cultural maintenance increases perceived discrimination and subsequent psychological distress among Turkish migrants in respective host countries. However, these results contradict with the findings from previous studies reporting that the strong identification with the heritage culture increases resilience against mental health problems (Pascoe & Richman, 2009). The contradictory findings might indicate that acculturation is a contextual factor and whether it is a protective factor or a risk enhancement for mental health and suicidal behaviours depend on the changing socio-political dynamics of the host-countries (Eylem et al., 2019). It is also important to know whether the mainstream culture adoption or the cultural continuity is important in migrants` lifestyle. Furthermore, the extent of the culture conflict is relevant. In light of the culture conflict theory, it could be argued that the greater the disparity between the demands of the host culture and the minority culture, the greater the risk for psychological distress, subsequent mental health problems and suicidal behaviours (Bhugra, 2004). With respect to the role of attachment, it could be that the insecure attachment styles, which are characterised by a sense of unworthiness and/or unlovability (e.g. Bowlby, 1973), are more likely to precipitate the anticipation of rejection from other General discussion ― 239 ethnic groups. Arguably, participating in daily social interactions with other minority and majority communities might provide them with positive intergroup contacts which may contradict the anticipation of rejection and reduce the risk for suicidal thinking and attempts in return (Eylem et al., 2019).

Cultural meaning of suicide and help-seeking for suicide The existing literature highlights the important role of the cultural meaning of suicide in relation to the elevated risk of suicide in some ethnic groups compared to the others (Bhui, 2010a; Hjelmeland, 2010; 2011; Colucci, 2013). Yet, there is insufficient data on these cultural views of suicide among Turkish migrants (Eylem et al., 2016). Thus in Chapter 4, the cultural meaning of suicide among Turkish migrants was investigated through a cross-cultural qualitative study in the Netherlands and in the UK. The results indicated more similarities than differences in meanings attributed to suicide and help-seeking for suicide in both countries. Further, perceived failure of meeting responsibilities towards family and community was central to the main stressors (acculturation, transformation in family system) leading to suicidal behaviours. Additionally, feelings of shame (putting down the family) and the stigma associated with suicide were also identified as major 8 barriers limiting Turkish migrants’ access to their informal (i.e. friends and family) and formal (i.e. GP) networks to seek help (Eylem et al., 2016). The preceding results are consistent with other qualitative studies concerning the Turkish migrants in Germany (e.g. Heredia-Montesinos et al., 2018). They are also relevant to other populations sharing collectivistic cultural values such as Gypsy communities (Lester, 2015), indigenous American communities (e.g. Lester, 2012, p.71), Chinese communities (Tseng & Wu, 1985, p.15), African communities (Mugisha et al., 2012) and South Asian Muslim communities in the UK (Till & Bhugra, 2015; Kamal & Loewenthal, 2002). The findings of Chapter 4 have also called attention to the shame and stigma associated with suicide restricting access to the informal (e.g. relatives and friends) and formal (e.g.GPs) networks during the help-seeking process. This finding contradicts with 240 ― Chapter 8 the cross-cultural research highlighting more accepting and helping reactions to suicide disclosures among Turkish young adults compared to a number of European populations (Slovakia, Switzerland, Austria) (Eskin, 1999a;1999b; Eskin et al., 2010; Eskin et al., 2015). It could be arged that the reactions of the informal network to suicide disclosure might be perceived as patronising and/or rejecting and this would exacerbate the feelings of shame and might eventually lead to not to seek help (Eylem et al., 2016). For example, observations about people`s attempt to keep suicidal person occupied with fun activities may look like helping reactions but they may in fact block the communication about real issues causing suicidal thinking (Eylem et al., 2016).

Cultural continuity One of the important research questions based on the findings of the Chapter 4 is whether there is continuity (or change) in the social and cultural factors contributing to suicidal behaviour among Turkish populations in different contexts. The answer to this question remains unclear as there is no comparison group from mainland Turkey in none of the existing studies (Eylem et al., 2016; Heredia-Montesinos et al., 2018). This limits the conclusions about the cultural views on suicide and how these views are maintained and/or re-constructed in different contexts. Central to addressing this issue, Chapter 5 presented a systematic review comparing the characteristics and the precipitating factors for suicidal behaviours of Turkish migrant women between countries of settlement and country of origin. There was a consistency in social stressors of suicide (e.g. marital conflict, domestic violence and honour related violence) across the studies in Turkey and in Europe. Overall, these results point to a possible link between Turkish women`s suicidal behaviour and violence against them in the context of family and/or intimate relationships with a spouse. Further support for this finding comes from cross-cultural studies indicating that domestic violence is crucial to the emergence of suicidal behaviours especially among women of ethnic minority, migrant and/or refugee backgrounds (Devries et al., 2011; Colucci & Heredia-Montesinos, 2013; Canetto, 2015). General discussion ― 241

In light of the preceding evidence indicating the elevated risk of suicidal behaviours among Turkish migrants, an important clinical implication is to make psychological services accessible during the help-seeking process. One of the major limitations of the available interventions for suicidal behaviours is that they rely on face-to-face contact with a practitioner, therefore they may not be accessible for Turkish migrants, mainly due to social (e.g. stigma-related), and organisational (e.g. waiting times and service availability, cultural mismatch between service provider and user) barriers (Jacobs et al., 2011; Department of Health, 2017; Institute of Medicine, 2002; Leigh & Flatt, 2015; Poppleton & Gire, 2017). It is suggested that one promising way to tackle these barriers is through culturally adapted e-mental health interventions (Cuijpers et al., 2019). Specifically, it is relatively easy to offer treatments through PCs, mobile phones or tablets in users preferred language and cultural values without time restriction (Eylem et al., 2015; Cuijpers et al., 2019). In order to explore this option, an existing e-mental health intervention for managing suicidal thoughts has been adapted according to the Turkish migrants in the Netherlands and in the UK. The following section discusses the results of the feasibility study and whether there is an indication for the effectiveness of the culturally adapted intervention compared to the information page (treatment as usual). 8

Culturally adapted e-mental health for managing suicidal thoughts among Turkish migrants Chapter 6 includes a description of a study protocol of an RCT aiming to investigate the effectiveness of a culturally adapted e-mental health intervention compared to a waitlist control group in reducing suicidal ideation among Turkish migrants in the Netherlands and in the UK. The e-mental health intervention has been developed by van Spijker and colleagues (van Spijker et al., 2010) and its effectiveness has been evaluated among general populations in the Netherlands (van Spijker et al., 2014), in Australia (van Spijker et al., 2018) and in Belgium (De Jaegere et al., 2019). Although a recent meta-analysis indicate the feasibility and efficacy of e-mental health interventions for treating suicidal ideation specifically (Arshad et al., 2019), the evidence for their effectiveness in ethnic minorities is 242 ― Chapter 8 not known. Thus, the existing e-mental health intervention of van Spijker and colleagues (van Spijker et al., 2010), was adapted linguistically and also culturally in terms of the use of cultural idioms of distress, cultural case examples representing the target group with similar problems. Further, adaptations in the use of the concepts and some theoretical modifications have also been made based on the qualitative study which was described in Chapter 4 (see Appendix A). The feasibility of the adapted intervention was subsequently evaluated. Additionally, the effectiveness of the intervention in comparison to the wait-list control group was investigated and reported in Chapter 7. Results pointed to the possible implementation issues restricting its usability in daily life. More specifically, not feeling connected with the intervention because of the emphasis on suicidal thoughts and /or the overrepresentation of more traditional cultural values were highlighted as potential barriers. Arguably, the adaptations were overgeneralised, thereby were not representative of all the sub-groups within the Turkish diaspora in the UK and in the Netherlands. Additionally, there was no evidence that the intervention group lead to better health outcomes than the control group. In both groups there was a strong indication for the reduction in suicidal ideation, depression and hopelessness, but not for worrying and quality of life. One of the important implications of the findings reported in Chapter 7 is the safety of the e-mental health intervention. Even though there were participants with high levels of suicidal thoughts, depression and worrying, the safety protocol had not been activated. This finding is in line with other trials investigating its effectiveness in general populations in the Netherlands (van Spijker et al., 2014), in Belgium (De Jaegere et al., 2019) and in Australia (van Spijker et al., 2018). If anything, our results suggest that vulnerable groups such as minority groups with severe suicidal thoughts and depression could be studied safely. The finding that there was no between-group differences in suicidal ideation scores, as might be expected in a small feasibility study, differed from the growing evidence supporting the effectiveness of e-mental interventions in reducing suicidal thinking (van Spijker et al., 2014; Arshad et al., 2019; De Jaegere et al., 2019). It is however, in line with General discussion ― 243 a more recent trial of van Spijker and colleagues (van Spijker et al., 2018) which did not indicate a significant effect of the intervention in reducing suicidal ideation in Australia (van Spijker et al., 2018). We argue that these contradictory findings might point to the attitudinal (e.g. stigma), geopolitical (e.g. accessibility to internet) or cultural barriers (e.g. cultural or religious prohibitions against suicide and help-seeking) which might vary between countries and between ethnic groups within countries. Such differences might impact the implementation of the e-health intervention in different contexts. The reasons for the low access threshold of our intervention remains unclear. It is possible that the strong stigma attached to suicide and fear of disclosing ones’ identity during the help-seeking process, which were discussed in Chapter 4 extensively, were some of the barriers at the individual level. Possible low mental health literacy about suicidal thoughts could have been another factor compromising the relevance of the intervention. At the system level, some cultural adaptations (e.g. cultural case examples representing more traditional cultural values) might have restricted the implementation of the intervention in daily life (Jacobs et al., 2011).

8 limitations The results presented in this thesis should be viewed in light of several important limitations. First of all, the concepts race, ethnicity and minority and majority statuses were oversimplified. For instance in Chapter 2, the mutually exclusive category approach (i.e. re-allocating individuals in existing categories defined by the previous research) was used in order to categorise the ethnic and religious groups in the included studies (Conelly et al., 2016). We also defined minority and majority statuses based on the country of the included studies. This approach was chosen in order to facilitate the comparability of the studies across countries (Conelly et al., 2016). However, we were not able to capture the specific ethnic groups within the broad racial classifications (e.g. Asian, Black, White) in the included studies. An alternative to the preceding approach is the multiple characteristics approach taking into account the various aspects of ethnicity such as language, country of birth, nationality and religiosity (Conelly et al., 2016). 244 ― Chapter 8

The latter offers more effective approach to the measurement of ethnicity in detail but it is not pragmatic, therefore restricts the comparability of different ethnic and/or racial categories (Conelly et al., 2016). Defining the preceding concepts is complicated and should include many factors including history, religion, language which tend to change depending on the socio-political dynamics (Conelly et al., 2016). We did attempt to make a distinction among Turkish-speaking migrants on the basis of such socio-political dynamics (e.g. Kurdish, Alevi, Turksih Cypriot etc.) in Chapter 4. However, there was a small number of participants from each ethnocultural groups in the sample. Additionally, no distinctions were made based on the history of the included participants in Chapter 3, 4 and 7. Overall, these limitations restrict the generalisability of the findings to the entire ethnic minority populations in general and to the Turkish-speaking migrant populations in Europe specifically. Second, the concepts of stigma and acculturation were multidimensional and were not clearly defined in the literature. For instance, in the stigma literature, different types of stigma are defined such as treatment stigma and mental illness stigma. Suicide stigma has been receiving attention only recently in the literature (Rimkeviciene et al., 2019). Since suicide stigma has yet to be investigated, we were not able to investigate its impact among ethnic minorities relative to majorities in this thesis. We argue that although conceptually different, different types of stigma might have a similar underlying mechanism and therefore furthering our understanding of mental illness stigma would have an implication on the impact of different types of stigma in general (Fox et al., 2018). Further, difficulties with operationalising definitions may have potentially hindered the measurement of the key concepts in this thesis. In Chapter 2, the variety of outcomes (self-report questionnaires and vignettes) may have contributed to the high heterogeneity of the results found on racial differences in mental illness stigma. Most of the time, these outcomes were developed for the purposes of the study, and were not standardised. In Chapter 2, limiting the analyses of the studies to the ones which used standardised outcomes, reduced the heterogeneity of the results indicating the lack of clarity in the stigma literature. Similarly, in Chapter 3, we used the adapted version of the acculturation General discussion ― 245 scale which was shown to be reliable in our study (Eylem et al., 2019) and also in previous research (Ünlü İnce et al., 2014a). Even though the new items were reliable, their cross- cultural validity were not tested. Third, there were methodological limitations with the designs of the studies in Chapter 3 and 4. Based on the results of Chapter 3, we can only suggest tentative causal inferences due to the cross-sectional design. Future research should utilise prospective designs in order to test the proposed relationships. Further, a convenience sampling method was used in Chapters 3 and 4 and this restricts the representativeness of the sample. Because of this sampling method, self-selection bias was highly likely in both studies. Another important limitation in the study reported in Chapter 4 was the memory bias. The interview data relies on the participants` retrospective accounts of the events and memories. Memory bias and selection bias were also relevant in the study reported in Chapter 5 which relied on the hospital and police records of Turkish women who had attempted and/or completed suicide. These reports are also based on the selective memories of individuals. Another methodological limitation related to the design of the study was reported in Chapter 7. Participants had access to other sources of information and/or usual care 8 (e.g. GP, psychologist) throughout their participation in the trial. Although there were no differences between groups at baseline in terms of the help-seeking, we did not carefully monitor their exposure to the available service and/or information during the study. It is therefore possible that receiving help and/or exposure to other information confounded the results which might have contributed the improvements in both groups. Fourth, the poor quality of the studies included in the systematic reviews in Chapter 2 and 5 were also notable. In Chapter 2, studies of poor quality had higher effect sizes. In Chapter 5, even though the studies included from Turkey and Europe pointed to the consistency of gendered factors such as “honour-related” violence, these factors were often not empirically tested and were only speculative in the discussions (van Bergen et al., 2019). Furthermore, in chapter 5 there was a possible variability of data collection and reporting across the countries and across the regions within Turkey. Even though the 246 ― Chapter 8 quality assessment of the included studies (see table 1 in chapter 5) were mainly consistent, there was indeed some selective reporting and assessment. For instance, even though the “honour-related” violence was not frequently reported in the included studies, it was extensively reported in a community study from a traditional area in Turkey (KAMER, 2011). Similarly, there was low report of mental illnesses in studies from traditional areas in Turkey. Overal, these limitations restrict the conclusions of the systematic reviews in this thesis. Finally, the intended number of participants to run the RCT was not reached in the time available for the completion of the study. Despite the use of recruitment methods such as facebook and public events, which have previously been shown to be successful (e.g. Ünlü İnce et al., 2014b), there was a low-access threshold of the intervention throughout the recruitment process. The small sample size also restricted the statistical analyses as we were not able to perform complex statistical procedures such as multiple imputation. implications for clinical practice The results of this thesis have led to several implications for clinical practice. First of all, the positive finding from the meta-analysis (Chapter 2) suggests that more attention should be paid to to tailor the existing anti-stigma interventions to the specific racial and/or ethnic groups (Gronholm et al., 2019). Given the propensity to experience stigma following suicidal thoughts in the Turkish migrant sample in Chapter 4 and 7, improving their mental health literacy should be a target for therapeutic goals. It is also important to inform the informal network (friends and relatives) about the risk factors, available resources for help and helpful ways to engage with people feeling suicidal (e.g. referring to professionals) through community-based suicide prevention campaigns. An example for an anti-stigma intervention is the partnership project between Derman, a Non-Governmental Organisation providing bi-lingual psychological and advocacy services to Turkish-speaking migrants, and the East London NHS Foundation Trust in London, UK. This projects concerns a community-based anti-stigma campaign for General discussion ― 247 improving suicide literacy among the Turkish-speaking populations in London in the UK. The campaign includes liaising with the representative community organisations in organising psycho-educational meetings (i.e. how to see the signs for suicidal behaviours, how to talk to those who might be at risk, how to refer them to available services for further help) with the informal networks of those who might be at risk for suicidal behaviours. Furthermore, the assessment of acculturation and insecure attachment styles may be useful in identifying high risk profiles of migrants. Mental health professionals’ awareness of the acculturation experiences as potential risk or protective factors for suicidal behaviours should be promoted. This approach might help service providers to engage with the service users during their help-seeking process. In terms of the high risk profiles, clinical interventions should also target Turkish migrant women aged between 14 and 25 specifically. It is fundamental to investigate the presence of mental conditions such as depression, and it is equally important to question the presence of socio-cultural risk factors such as gender based violence and/or “honour- related” violence during the clinical assessment. With respect to the real world cultural adaptation practices, Chapter 7 emphasises 8 the importance of taking the diversity of the migrant populations into account. Overgeneralisation is one of the common challenges encountered whilst adapting interventions culturally according to particular populations. The results of the feasibility study described in Chapter 7 indicate that acculturation might be an important factor in making decisions about the level of cultural adaptatation. More specifically, cultural adaptations of the interventions should not exclusively represent the norms and values of the settled community but should also address the values of those who are in the process of settlement. Finally, the safety precautions for the implementation of the e-mental health amongst Turkish migrants with suicidal ideation have been considered in Chapter 7. It could be that providing the intervention as part of the support services offered by, for example, the 113 online in the Netherlands or the Samaritans in the UK might anonymise the 248 ― Chapter 8 help-seeking process and overcome potential barriers relating to the fear of disclosing ones’ identity. Additionally, the therapeutic effect of guidance and personal coach was emphasised. Future studies should incorporate more guidance and personal coaching in participants` preferred language in order to improve engagement. implications for fUtUre research This thesis has thrown up a number of important research questions. Chapter 2 shows that the intersectional impact of stigma could be explored further by examining the role of other possible moderators such as diagnoses of CMDs. Thus, the following questions could be investigated in future systematic reviews and meta-analyses: Among those who have received a diagnosis of a CMD, are there any differences in the experience of mental illness stigma between racial and/or ethnic minorities and majorities? Among racial and/ or ethnic minorities, are there differences in mental illness stigma between those who have been diagnosed with CMDs and those who have not? Another interesting area to examine could be whether there are differences in mental illness stigma between racial and/or ethnic minorities, migrants and racial and/or ethnic majorities. Chapter 3 pointed to the importance of investigating the differences in associations between acculturation and suicidal behaviours between first and later generations given that acculturation experiences may vary between generations. Further, in order to investigate the relevance of the cultural continuity thesis in suicide epidemiology, Chapter 4 highlighted the importance of including a comparison group from mainland Turkey. This is crucial to study whether the cultural meaning of suicide and help-seeking for suicide were maintained and/or re-constructed in different diasporas. Providing further insight on the preceding question about the cultural continuity argument, the results reported in Chapter 5 indicated that future studies could also investigate the extent and/or the influence of the ‘traditionality’ of the regions in Turkey. Since the Turkish migrants in Europe often migrate from the traditional regions of General discussion ― 249

Turkey, future research should investigate whether risk and precipitating factors of suicide amongst females are influenced by the ‘traditionality’ of the place of origin. Chapter 7 highlighted the challenges of engaging with Turkish migrants during the recruitment process for e-mental health interventions (Eylem et al., 2019; Ünlü İnce et al., 2014b). The principles of “task shifting” or “task-sharing” could be relevant to improving engagement. According to the WHO, task shifting is “the rational redistribution of tasks among health workforce teams” (WHO, 2008). This refers to the specific functions that are shifted from highly qualified health workers to non-professional or lay health workers with shorter training and fewer qualifications. It could be that liaising with community centres might generate an institutional barrier especially among sub-groups of migrants who do not feel connected with their own ethnic group and /or specific community centres representing a particular political ideology and/or ethnocultural group (Bhui, 2010a). This barrier could be relevant to the diverse transnational migrant populations such as the Turkish migrants in Europe (Eylem et al., 2016). Thus, trained lay people, who are representative of the various sub-groups and are not based in community centres might be helpful in removing such barriers. In mental health research, task shifting has been used in several well-designed large trials (Patel et al., 2010; 2017; Rahman, et al., 8 2008; 2016; Bryant et al., 2017). In these trials, non-professional counsellors without a background in mental health care, are locally recruited and have received training to deliver psychological treatments and care management to people with CMDs. These trials found positive effects, for depression, although the number of studies were relatively small (Joshi et al., 2014; van Ginneken et al., 2013). Future research should investigate whether using trained lay people is feasible and effective for engaging with Turkish migrants and other ethnic minorities in the recruitment process for suicide research. Further, the results of Chapter 7 highlight the importance of investigating effective cultural adaptation practices further. Additionally, it appears that some cultural adaptation practices might overlook individual differences and stereotype specific cultural values and norms (Rathod et al., 2015). Thus, future research should also investigate whether some cultural adaptation practices might have adverse effects for the members 250 ― Chapter 8 of a particular ethnic group. An important implication of investigating the potential benefits and harms of real world cultural adaptations, will be to refine the existing cultural adaptation frameworks with the intention of standardising current cultural adaptation practices. Culturally adapted e-mental health interventions are promising in terms of improving engagement and effectiveness of the psychological interventions. However, one of the important messages based on the results of Chapter 7 is that, more feasibility and exploratory trials are needed before moving onto effectiveness testing of the culturally adapted e-mental health interventions in complex settings. In this thesis, the lack of effect of the culturally adapted intervention pointed to the implementation issues. Within the framework of the Medical Research Council (MRC) (Craig et al., 2008), it could be argued that lack of effect of complex interventions might not reflect a genuine ineffectiveness. Thus, a thorough process evaluation is needed for further research in order to detect implementation problems related with culturally adapted e-mental health interventions. With respect to the diversity of the migrant and ethnic minority populations (e.g. variation in level of acculturation), larger sample sizes may take accout for the variability in future studies (Craig et al., 2008). In terms of the appropriate designs for the effectiveness testing of the culturally adapted e-mental health interventions, preference trials and randomised consent trials could be considered for future studies. Since there were participants who preferred not to be allocated to a wait-list group in Chapter 7, it could be that basing treatment allocation on participants` preferences could improve the engagement (Craig et al., 2008). Additionally, including another arm offering a face-to-face treatment within the design of future studies could make the participation more appealing to those who have strong preferences for face-to-face treatment over e-mental health. Finally, to reduce potential barriers such as low mental health literacy and stigma, future studies could employ a ‘two steps approach’ in which ethnic minorities receive a low intensity psycho-educational intervention as a first line intervention followed by a subsequent higher intensity intervention for those who have severe suicidal ideation General discussion ― 251 and depression symptoms. There are examples of such low-intensity community based psycho-educational initiatives in engaging with Hispanic populations in the US (Caplan, 2016) and Orthodox Jewish communities in London, UK (Perry et al., 2018) during their help-seeking process. Lastly, Chapter 7 highlights the importance of investigating whether there is an association between the intensity of suicidal ideation, stigma and help-seeking among Turkish migrants. conclUsion This thesis provides insight into the impact of stigma related to mental health for racial minorities in general and Turkish migrants specifically. It also develops the current understanding of the socio-cultural context related with the suicidal behaviours among the Turkish migrants in Europe. Most important of all, the thesis presents one of the first studies to have specifically focused on the feasibility of a culturally adapted e-mental health intervention for suicidal ideation among Turkish migrants presenting an increased risk for suicidal behaviours. An important message of this thesis is that suicidal behaviours of Turkish migrants should be understood in relation to structural inequalities and power imbalances in areas such as gender role expectations, which exist in both 8 pre- and post-migration contexts. Additionally, stigma related to mental health and suicide highlights the importance of facilitating the timely access to information, services and support during their help-seeking process. Moreover, the challenges experienced whilst implementing the e-mental health intervention in this thesis suggest that adapting interventions culturally is necessary but not sufficient to optimise these interventions for ethnic minority and migrant populations with complex needs. Finally, this thesis should offer a road map to investigate whether online interventions could provide a feasible and an effective alternative among underserved populations including ethnic minority groups who are at risk for suicidal behaviours.

9

SUMMARY 254 ― Chapter 9

uicide is a global public health problem with enormous consequences at individual and societal levels. Some ethnic groups worldwide are more at risk for suicidal behaviours and in addition to the elevated risk, there is an unequal access to mental health services among ethnic minorities. Thus, Sethnic minorities are considered to be more vulnerable to suicidal behaviours and are now being prioritised in national suicide prevention strategies in many countries. Turkish migrants in Europe have been identified in the literature as a high risk group for suicidal behaviours. They present an increased risk for suicidal behaviours when compared with their native counterparts from the respective host countries. They also present cross-national consistency in suicide epidemiology when compared with their counterparts from their country of origin (Turkey). There might be an interplay between socio-cultural factors such as migration history, gendered factors such as “honour- related” violence (i.e. autonomy restrictions among women) and psychological factors such as hopelessness, generating an increased risk for suicidal behaviours among Turkish migrants. Recently, e-mental health interventions have been shown to be an effective and feasible treatment option for suicidal ideation specifically. However, evidence for their feasibility and effectiveness for other ethnic groups is currently lacking. Thus, there is an urgent need to further investigate whether the current provision of psychological services for migrants and ethnic minorities could be improved through online interventions and whether culturally adapted e-mental health interventions are feasible and effective options in the treatment of suicidal ideation among ethnic minorities. Based on this knowledge gap, it is hoped that this thesis will provide further insight into the socio- cultural risk factors and characteristics of suicidal behaviours among Turkish migrants in Europe. It may also inform the applicability of culturally relevant e-mental health interventions for Turkish migrants with suicidal ideation. In Chapter 2, the impact of mental illness stigma was assessed by examining the racial differences in mental illness stigma for CMDs between racial minorities and majorities. The moderating effect of certain study characteristics such as race, quality of the studies Summary ― 255 and type of stigma outcomes (vignette vs self-report) were also investigated. Results showed more mental illness stigma for CMDs among racial minorities compared to majorities. There was a possible moderating effect of studies with high risk of bias but no moderating effects were found for race or type of stigma outcomes in multivariate analyses. Chapters 3, 4 and 5 focused on the cultural and psychosocial characteristics that increased the likelihood of a suicidal crisis (such as socio-demographic factors or mental illness) and precipitating factors (such as perceived causes or domestic violence) for suicidal behaviours among the Turkish populations in Europe and in Turkey. Chapter 3 investigated the relationship between acculturation and suicidal ideation amongst the Turkish migrants in the Netherlands through a cross-sectional survey design. It also investigated the mediating effect of hopelessness and the moderating effect of secure attachment. Higher participation in the social life of the host culture was associated with less hopelessness and less suicidal ideation. Greater maintenance of one’s culture of origin was associated with higher hopelessness and higher suicidal ideation. More participation was associated with less suicidal ideation particularly amongst those with less secure attachment styles. In Chapter 4, the cultural meaning of suicide was investigated through focus group 9 and one-to-one interviews with lay people and professionals representing the Turkish- speaking populations in the Netherlands and the UK. There were more similarities than differences in the narratives of the participants in both countries. Failing in their responsibilities towards family and community was central to the main stressors (acculturation, transformation in family system) leading to a greater likelihood for suicidal behaviours. In the help-seeking process for suicide, feelings of shame (reflecting badly on the family) and the stigma associated with suicide were identified as major barriers limiting Turkish migrants’ access to their informal (i.e. friends and family) and formal (i.e. GP) support networks to seek help. Chapter 5 examined the precipitating factors and characteristics for suicide and attempted suicide for the Turkish women in Turkey and in Europe. There were more 256 ― Chapter 9 similarities than differences in characteristics and precipitating factors for suicidal behaviours between the Turkish women in Turkey and the Turkish migrant women in Europe. The results highlighted the presence of mental illness, notably depression and cultural continuity of psychosocial risk factors such as patriarchy and gendered factors (i.e. gendered based violence, autonomy restrictions) leading to an elevated risk of suicide and attempted suicide among Turkish women in different contexts. Chapters 6 and 7 presented a RCT assessing the feasibility and indications for effectiveness of a guided e-mental health intervention for Turkish migrants with suicidal ideation. The intervention was based on the principles of the CBT. It was then culturally adapted according to the Turkish migrants in the UK and the Netherlands based on focus groups and one-to-one interviews with the members of the community and the professionals working with the community. The research protocol was described in Chapter 6. Chapter 7 reported on the results of the feasibility and a pilot RCT showing no indication that the intervention group lead to better health outcomes than the control group. In both groups there was a reduction in suicidal ideation, depression and hopelessness, but not in worrying and quality of life. The low scores on usability of the intervention pointed to the implementation issues compromising its usability in daily life by the target population. The culturally adapted content was viewed by the majority of the participants as relevant, appropriate and acceptable. Participants with mild suicidal ideation specifically did not feel connected with the intervention for a number of possible reasons:1) Because the emphasis of the intervention was on suicidal thoughts, and 2) because of the overrepresentation of the traditional Turkish cultural values such as “honour”, in the content of the intervention. The major strength of the study was that the safety protocol had not been activated due to the risk of a suicide attempt during the study. Finally, the barriers identified during the process such as not feeling connected with the intervention, suggests that a number of improvements with the design of the study and with the intervention itself could be made before moving onto further effectiveness testing. Summary ― 257

9

9

SAMENVATTING (SUMMARY IN DUTCH) 260 ― Chapter 9

uïcidaliteit is een wereldwijd probleem voor de volksgezondheid met enorme gevolgen voor het individu en de maatschappij. Sommige etnische groepen hebben een verhoogd risico op suïcidaal gedrag. Naast een verhoogd risico op suïcidaal gedrag is het zorggebruik onder etnische minderheden laag. SMede door deze verminderde toegang tot zorg worden etnische minderheden gezien als kwetsbare groep. Suïcidepreventie onder etnische minderheden krijgt daarom in veel landen prioriteit bij nationaal suïcidepreventie beleid. In de literatuur worden Turkse migranten in Europa geïdentificeerd als een risicogroep voor suïcide. In vergelijking met de autochtone bevolking heeft deze groep een verhoogd risico op suïcidaal gedrag. Tevens is er internationaal consistentie te zien ten aanzien van de epidemiologie, want ook in Turkije bestaat er een hoog risico op suïcidaal gedrag onder sommige groepen in de bevolking. Er is dus mogelijk sprake van een wisselwerking tussen sociaal-culturele factoren, zoals migratiegeschiedenis, gender gerelateerde factoren zoals eer gerelateerd geweld (bijvoorbeeld autonomiebeperkingen bij vrouwen), en psychologische factoren, zoals hopeloosheid, die zorgen voor een verhoogd risico op suïcidaal gedrag bij Turkse migranten. Onlangs is aangetoond dat e-health interventies effectief zijn in het behandelen van suïcidale gedachten en dat een online setting geschikt is om deze gedachten aan te pakken. Op dit moment is er echter nog geen bewijs voor de haalbaarheid en effectiviteit om specifieke etnische groepen ook online te behandelen met dergelijke interventies. Het is dus van belang om verder te onderzoeken of het huidige aanbod van psychologische interventies voor migranten en etnische minderheden kan worden verbeterd. Specifiek wordt onderzocht of het implementeren van e-health interventies en of cultureel aangepaste e-health interventies haalbare en effectieve opties zijn om suïcidale gedachten bij etnische minderheden te behandelen. Vanwege een gebrek aan wetenschappelijk onderzoek is het doel van dit proefschrift om meer inzicht te verschaffen in de sociaal- culturele risicofactoren en kenmerken van suïcidaal gedrag bij Turkse migranten in Europa. In het proefschrift wordt ook de toepasbaarheid van cultureel relevante e-health interventies voor Turkse migranten met suïcidale gedachten onderzocht. Samenvatting ― 261

In Hoofdstuk 2 van het proefschrift is de impact van stigma voor psychische aandoeningen onder Turkse migranten in Nederland en Engeland beschreven. Dit werd gedaan door de verschillen in stigma voor veel voorkomende psychische aandoeningen tussen etnische minderheden en de autochtone bevolking te onderzoeken. Hierbij werden modererende effecten onderzocht zoals etniciteit, de kwaliteit van de studies en type stigma-uitkomsten (vignet versus zelfrapportage). De resultaten laten zien dat er meer stigma’s zijn voor veelvoorkomende psychische aandoeningen onder etnische minderheden in vergelijking met de autochtone bevolking. Er was een mogelijk modererend effect van studies met een hoog risico op bias, maar er werden geen modererende effecten gevonden voor etniciteit of type stigma-uitkomsten in een multivariate analyse. De hoofdstukken 3, 4 en 5 zijn gericht op het onderzoeken van de culturele en psychosociale kenmerken die de kans op suïcidaliteit vergroten (zoals sociaal- demografische factoren of psychische aandoeningen) en precipiterende factoren (zoals waargenomen oorzaken of huiselijk geweld) voor suïcidaal gedrag onder Turkse populaties in Europa en in Turkije. In hoofdstuk 3 werd de relatie tussen acculturatie en suïcidale gedachten bij Turkse migranten in Nederland cross-sectioneel onderzocht. Het mediërende effect van hopeloosheid en het modererende effect van veilige hechting werden tevens onderzocht. 9 Actievere deelname aan het sociale leven binnen de Nederlandse samenleving werd geassocieerd met minder hopeloosheid en minder suïcidale gedachten. Een sterker behoud van de cultuur van oorsprong werd geassocieerd met meer hopeloosheid en meer suïcide gedachten. Vooral onder individuen met minder veilige hechtingsstijlen, was meer deelname aan het sociale leven binnen de Nederlandse samenleving geassocieerd met minder suïcidale gedachten. In hoofdstuk 4 wordt de culturele betekenis van suïcide onderzocht door middel van focusgroepen en individuele interviews met zowel leken als professionals die de Turkssprekende bevolking in Nederland en het Verenigd Koninkrijk (VK) vertegenwoordigen. Er waren meer overeenkomsten dan verschillen in de verhalen afkomstig uit beide landen. Het idee niet te voldoen aan verantwoordelijkheden voor 262 ― Chapter 9 familie en de gemeenschap was een centraal onderdeel van de belangrijkste stressoren (acculturatie, transformatie in het familiesysteem), wat door professionals werd geassocieerd met een grotere kans op suïcidaal gedrag. In het hulpzoekproces voor suïcidaliteit werden gevoelens van schaamte (een slechte uitstraling op het familie) en het stigma geassocieerd met suïcide, geïdentificeerd als belangrijke barrières. Deze barrières belemmeren de Turkse migranten om hulp te zoeken bij hun informele (bijvoorbeeld vrienden en familie) en formele (bijvoorbeeld de huisarts) zorg. In hoofdstuk 5 worden precipiterende factoren en kenmerken onderzocht behorende bij suïcide en pogingen tot suïcide onder Turkse vrouwen in Turkije en Europa. Er waren meer overeenkomsten dan verschillen in kenmerken en belangrijke factoren die samenhangen met suïcidaal gedrag tussen Turkse vrouwen in Turkije en Turkse migrantenvrouwen in Europa. De resultaten benadrukken de aanwezigheid van psychische aandoeningen, met name depressie, en de culturele continuïteit van psychosociale risicofactoren zoals het patriarchaat en gender-gerelateerde factoren (bijvoorbeeld huiselijk geweld en autonomiebeperkingen) die leiden tot een verhoogd risico op suïcide en poging tot suïcide onder Turkse vrouwen in verschillende contexten. De hoofdstukken 6 en 7 richten zich op een pilot randomized controlled trial (RCT) waarin de haalbaarheid en indicaties voor de effectiviteit van een begeleide e-health interventie voor Turkse migranten met suïcidale gedachten werd onderzocht in Nederland en het VK. De interventie is gebaseerd op de principes van cognitieve gedragstherapie. Het werd vervolgens cultureel aangepast voor de Turkse migranten in het VK en Nederland op basis van focusgroepen en persoonlijke interviews met leden van de gemeenschap en professionals die met de gemeenschap werken. Het onderzoeksprotocol wordt beschreven in hoofdstuk 6. In hoofdstuk 7 worden de resultaten beschreven van een pilot RCT. In de RCT werden na de e-health interventie geen betere gezondheidsresultaten gevonden voor de interventiegroep in vergelijking met de controlegroep. In beide groepen was er een afname van suïcide gedachten, depressie en hopeloosheid, maar geen afname van zorgen en kwaliteit van leven. De lage scores op bruikbaarheid van de interventie wezen Samenvatting ― 263 op de implementatieproblemen, waardoor de bruikbaarheid in het dagelijks leven van de doelpopulatie nog onduidelijk is. Wel werd de cultureel aangepaste inhoud door de meerderheid van de deelnemers beschouwd als relevant, passend en acceptabel. De deelnemers met milde suïcidegedachten vonden dat de interventie niet aansloot om een aantal redenen: 1) de nadruk van de interventie lag op suïcidale gedachten en 2) vanwege de oververtegenwoordiging van traditionele Turkse culturele waarden, zoals eer, in de inhoud van de interventie. Een positief punt van het onderzoek was dat het veiligheidsprotocol niet geactiveerd hoefde te worden, ondanks het risico op suïcide tijdens het onderzoek. De tijdens het onderzoek geïdentificeerde belemmeringen, zoals zich niet verbonden voelen met de interventie, suggereert dat een aantal verbeteringen met name in de opzet van het onderzoek en met de interventie zelf kunnen worden aangebracht voordat wordt overgegaan op verder effectenonderzoek.

9

9

ÖZet (sUmmary in tUrkish) 266 ― Chapter 9

ntihar kişisel ve toplumsal boyutta çok ağır sonuçları olan küresel bir halk sağlığı sorunudur. Dünya genelinde bazı etnik gruplar intihar davranışları riski altındadırlar ve buna ek olarak ruh sağlığı servislerine erişimleri kısıtlıdır. Bu nedenle, günümüzde birçok ülkede etnik azınlıklar intihar davranışlarına yatkın İolarak değerlendirilmekte ve kendilerine ulusal intiharı önleme startejilerinde öncelik verilmektedir. Avrupadaki Türkiyeli göçmenler literatürde intihar davranışları riski altında olarak tanımlanmaktadır. Göç ettikleri ülkelerdeki yerlilere kıyasla intihar davranışlarına yönelik yüksek bir risk göstermektedirler. Ayrıca Türkiye`de yaşayan benzerleriyle karşılaştırıldıkları zaman intihar verilerinde uluslararası tutarlılıklar gözlemlenmektedir. Türk göçmenler arasında intihar davranışları riskini güçlendirecek göç tarihçesi gibi sosyo-kültürel faktörler, şerefle ilgili şiddet (ör. Kadınlar arasında özgürlüğün kısıtlanması) gibi cinsiyetle ilgili faktörler ve ümitsizlik gibi psikolojik faktörler arasında etkileşim olabilir. Son zamanlarda internet üzerinden sağlanan psikolojik terapilerin özellikle intihar düşüncelerini tedavi etmede etkili ve uygulanabilir olduğu görülmektedir. Fakat bu terapilerin etnik azınlıklarda uygulanabilirliği ve etkililiği henüz kanıtlanmamıştır. Bu sebeple, internet terapilerinin etnik azınlıklar için olan psikolojik servislerin sağlanmasını güçlendirip güçlendirmediğine ve kültüre uyarlanmış internet terapilerinin etnik azınlıklarda intihar düşüncelerinin tedavisinde uygulanabilir ve etkili bir alternatif olup olmadığına yönelik daha fazla araştırma yapılmasına acilen ihtiyaç vardır. Bu bilgi eksikliğine dayanarak bu tezin Avrupa`da yaşayan Türk göçmenler arasında intihar davranışlarına sebep olan karakteristik özellikler ve risk faktörleriyle ilgili daha fazla öngörü sağlayacağı ümit edilmiştir. Ayrıca Türk göçmenler için kültüre uyarlanmış, intihar düşüncelerini tedavi etmeye yönelik olan internet terapisinin uygulanabilirliğiyle ilgili bilgi sağlaması amaçlanmıştır. İkinci bölümde ruh sağlığıyla ilgili tabuların sık rastlanan ruh sağlığı problemleri üzerindeki etkisi, ırksal azınlıklar ve çoğunlukları karşılaştırarak değerlendirildi. Ayrıca Özet ― 267

ırk gibi, araştırmalara yönelik bazı karakteristik özelliklerin, araştırmaların kalitesi ve ruh sağlığıyla ilgili tabuların sonuç tiplerinin (skeç veya öz bildirim ölçeği) moderatör etkisi araştırıldı. Sonuçlar ırksal azınlık gruplarına mensup bireylerde, ırksal çoğunluk gruplarına mensup bireylere kıyasla sık rastlanan ruh sağlığıyla ilgili daha çok tabuların olduğunu saptadı. Çok değişkenli analizlerde yanlılığı yüksek olan araştırmaların olası moderatör etkisi bulundu. Fakat ırk veya ruh sağlığıyla ilgili sonuç tiplerinin herhangi bir moderatör etkisi bulunmadı. Üçüncü, dördüncü ve beşinci bölümler Avrupa ve Türkiye`de yaşayan Türk toplumlarında intihar davranışlarının olasılığını artıran kültürel ve psikososyal faktörler (ör, sosyo-demografik özellikler veya ruh sağlığı sorunları) ve intihar davranışlarını tetikleyen faktörler (ör. Algılanan sebepler ve aile içi şiddet) üzerine odaklanıyor. Üçüncü bölüm Hollanda`da yaşayan Türk göçmenler arasında kültürleşme ve intihar düşünceleri arasındaki ilişkiyi bir anket çalışmasıyla inceledi. Ayrıca bu bölümde ümitsizliğin arabulucu etkisi ve güvenli bağlanma tipinin moderatör etkisi araştırıldı. Yaşanılan ülkedeki sosyal hayata daha yüksek katılımın, daha düşük ümitsizlik ve daha düşük intihar düşünceleriyle ilişkili olduğu bulundu. Kendi kültürel kökenini devam ettirmenin daha yüksek ümitsizlik ve daha yüksek intihar düşünceleriyle ilişkili olduğu bulundu. Özellikle güvensiz bağlanma tipine mensup kişilerde yaşanılan toplumun sosyal 9 hayatına daha fazla katılım, daha düşük intihar düşünceleriyle ilişkiliydi. Dördüncü bölümde intiharın anlamı, Hollanda ve İngiltere`de yaşayan Türk toplumları temsil eden sıradan kimseler ve uzmanlarla yapılan odak grubu toplantıları ve birebir görüşmelerle araştırıldı. Her iki ülkeden katılan kişilerin hikayelerinde farklılıklardan daha çok benzerlikler saptandı. Aile ve topluma karşı sorumlulukları yerine getirmedeki başarısızlık, intihar davranışlarına yol açan stres faktörlerini (ör. kültürleşme, aile sistemindeki değişiklik) tetikleyen en önemli sebepti. İntiharla ilgili yardım alma sürecinde, utanç duygusu (ailenin yüzünü kara çıkarma) ve intiharla ilişkili tabular, Türk göçmenlerin resmi olmayan (ör. arkadaş, aile) ve resmi (ör. aile doktoru) destek ağlarına erişimlerini önleyen en önemli bariyerler olarak belirlendi. 268 ― Chapter 9

Beşinci bölüm, Türkiye`de ve Avrupa`da yaşayan Türk kadınlar arasında intihara ve intihar teşebbüsüne yol açan tetikleyici faktörler ve karakteristik özellikleri inceledi. Türkiye`de ve Avrupa`da yaşayan Türk kadınlar arasında intihar davranışlarına yol açan tetikleyici faktörler ve karakteristik özellikler arasında farklılıklardan daha çok benzerlikler vardı. Sonuçlar; özellikle depresyon olmak üzere, ruh sağlığı sorunları yanında farklı bağlamlardaki Türk kadınlar arasında intihar ve intihara teşebbüs etme riskini artıran psiko-sosyal faktörlerde (ör. Cinsiyete bağlı şiddet, özgürlüğün kısıtlanması) kültürel bir devamlılık olduğunu gösterdi. Altıncı ve yedinci bölümler intihar düşünceleri olan Türkiyeli göçmenlere uyarlanmış internet terapisinin uygulanabilirliğine ve etkililiğine yönelik göstergeleri değerlendiren kontrollü bir deneysel çalışmayı sunmaktadır. Bahsedilen internet terapisi, Bilişsel Davranışçı Terapi ilkelerine dayanılarak geliştirilmiştir. Bu terapi daha sonra İngiltere ve Hollanda`da yaşayan Türkiyeli topluma mensup kişiler ve onlarla çalışan uzmanlarla yapılan odak grubu toplantıları ve birebir görüşmeler ışığında Türk kültürüne uyarlanmıştır. Bu araştırmanın protokolü altıncı bölümde tanımlanmıştır. Yedinci bölümde pilot deneysel kontrollü çalışma ve uygulanabilirlik çalışmalarının sonuçları sunulmuştur. Pilot çalışmada, deneysel grupta ve bekleme listesinde yer alan katılımcılar arasındaki intihar düşünceleri, ümitsizlik ve depresyon belirtilerinde azalma gözlemlenmiştir. Fakat bu sonuçlar endişelenme ve hayat kalitesiyle ilgili sonuçlar için geçerli değildir. İnternet terapisinin kullanılabilirliğiyle ilgili düşük puanlar, hedeflenen kesimin gündelik hayatlarında terapiyi kullanmalarını güçleştiren sorunlar olduğuna işaret etmektedir. Kültüre uyarlanmış internet terapisinin içeriği birçok katılımcı tarafından sorunlarıyla ilişkili, uygun ve kabul edilebilinir olarak değerlendirilmiştir. Özellikle düşük seviyede intihar düşünceleri olan katılımcılar internet terapisiyle şu sebeplerden dolayı kendilerini bağdaştıramadılar: 1) İnternet terapisinin intihar düşüncelerine odaklanması ve 2) İnternet terapisinin içeriğinde namus ve şeref gibi geleneksel Türk değer yargılarının çok fazla vurgulanması. Bu çalışmanın esas güçlü tarafı, araştırma süresi boyunca herhangi bir intihara teşebbüs etme riski yüzünden güvenlik protokolünün Özet ― 269 aktive edilmemiş olmasıdır. Son olarak, internet terapisiyle kendini bağdaştıramama gibi internet terapisini kullanım sürecinde belirlenen bariyerler araştırmanın tasarımında ve internet terapisinin kendisinde yapılması gereken düzenlemelere işaret etmektedir.

9

a

aboUt the aUthor 272 ― About the author

zlem was born and grew up in Cyprus and completed her bachelor in Psychology at the Teesside University in Middlesbrough in the UK, and completed her Masters degrees in Psychology (clinical emphasis) at the University of Nicosia in Cyprus and in Child and Adolescent ÖMental Health at the Institute of Psychiatry (IOP) Kings College University of London in the UK. In 2010 she worked on developing the self-harm questionnaire when she was based in the Belgrave Department, Kings College Hospital. She worked as a Visiting Research Associate whilst finalising her clinical and academic work at the Maudsley. Between 2010 and 2012 she worked with various Non-Governmental Organisations (e.g. Shoreditch SPA-Shoreditch Trust NHS, Netherlands Institute of Higher Education Ankara Turkey) and got involved in coordinating Youth in Action (YiA) projects in Cyprus, in Turkey and in the UK. In 2013 she started her PhD at the VU University Amsterdam aiming to investigate the feasibility and effectiveness of an e-mental health intervention for suicidal ideation among the Turkish migrants in the Netherlands and in the UK. She collaborated with the Centre for Psychiatry, Queen Mary University of London as a visiting research fellow throughout her PhD. In 2013 she was awarded with the full tuition scholarship by the Beck Institute Philadelphia, and with a personal grant by the European Commission. During her PhD, Ozlem developed an interest in public engagement as a means of engaging with diverse populations in their help-seeking process for suicidal behaviours. In collaboration with several community organisations, stakeholders and key individuals, she organised public events, aiming to raise awareness about suicide and issues around it in Turkish migrant populations. She has been awarded with public engagement grants and recognitions by the Centre for Mind in Society and the Centre for Public Engagement-Queen Mary University of London for her public engagement work. More recently, she has been awarded with the Emerald Publishing Real Impact Award, Highly Commended in Individual Impact Category for contributing to the real impact of her research through public engagement. Currently, Ozlem works as a project worker in a partnership project between Derman, a Non-Governmental Organisation providing bi-lingual psychological and advocacy services to Turkish, About the author ― 273

Kurdish and Turkish Cypriot migrants, and the East London NHS Trust in London in the UK. Her role includes developing and piloting psycho-educational awareness sessions aiming to improve suicide literacy and suicide related stigma in the Turkish- speaking communities in London. She is also liaising with the relevant professionals and community groups in promoting and running these sessions.

A

p

pUblications 276 ― Publications pUblications Cuijpers, P., Eylem, O., Karyotaki, E., Zhou, X., & Sijbrandij, M. (2019). Psychotherapy for depression and anxiety in low- and middle-income countries. In D. Stein, & J. Bass (Eds.), Global Mental Health and Psychotherapy: Adapting Psychotherapy for Low- and Middle-Income Countries (pp. 173-192). Elsevier. https:// doi.org/10.1016/b978-0-12-814932-4.00008-2 Eylem, O (2009). Which components of CBT are the most suitable for use in primary care? London Journal of Primary Care, Retrieved April 21, 2009, from http://www. londonjournalofprimarycare.org.uk/articles/ Eylem, O (2010). Development and Preliminary Validation of the Self-Harm Questionnaire (SHQ). LAP Lambert Academic Publishing AG & CO.KG. Eylem, O (2010). The Effects of Perfectionistic Self-Presentation on Social Anxiety. Scientific Conference: Psychology, Traditions and Perspectives (pp.1-8). Blagoevgrad/Bulgaria. Eylem, O (2012). Positive Parenting Programmes in Early Childhood Development: Implications for Turkey in Aarsen, J. and van Oudenhoven, N. (2012) New horizons. A fresh look at Early Childhood Development Turkey. Bernard van Leer Foundation, Utrecht, The Netherlands. Eylem, O., Dalgar, İ., Tok, F., Ünlü İnce, B., van Straten, A., Kerkhof, A.J.F.M., Bhui, K. (2019). The association between acculturation and suicidal ideation among Turkish migrants in the Netherlands. Psychiatry Research,275,71-77. Eylem, O., de Wit, L., van Straten, A. et al. (2020) Stigma for common mental disorders in racial minorities and majorities a systematic review and meta-analysis. BMC Public Health 20, 879 . https://doi.org/10.1186/s12889-020-08964-3 Eylem, O., van Bergen, D.D.,Rathod, S., van Straten, A.,Bhui, K., Kerkhof, A.J.F.M (2016). Canına Kıymak ‘crushing life energy’: A qualitative study on lay and professional understandings of suicide and help-seeking among Turkish migrants in the UK and in the Netherlands. International Journal of Culture and Mental Health, 9,1-16. Eylem, O.,van Straten, A., Bhui, K., Kerkhof, A.J.F.M (2015). Study protocol: Reducing suicidal thinking among Turkish migrants living in The Netherlands and in the UK: Effectiveness of an online intervention.International Review of Psychiatry,1,72-81. Publications ― 277

Kuester, L., Eylem, O., Melissourgaki, Z., Freestone, M., Dein, S., Bhui, K. (2019). Interventions for Mental Health Across Cultures: Post-conflict migration and gender- related mental health challenges. Inform Final Project Report. Centre for Protection of National Infrastructure. London, UK. Rathod, S., Persaud, A., Naeem, F., Pinninti, N., Tribe, R., Eylem, O., Pihiri, P., Muzaffar, S., Irfan, M. (2019). Culturally adapted interventions in mental health: Global position statement. World Cultural Psychiatry Research Review. Under review or in preparation Eylem, O., Eskin, M. (2020). Suicidal Behaviour and Social Cultural Factors Among Immigrants of Turkish Origin in Germany in Scholer-Ocak, M., Khan, Moosa M. (2020). Suicide Across Cultures. Oxford Cultural Psychiatry Series, UK. Eylem, O., Khaliq, A., Irshad, S., Husain, N. (2020). Identifying evidence of effectiveness and feasibility of the culturally adapted psychological interventions for ethnic minorities with suicidal behaviours in High and Middle Income Countries: A systematic literature review. Manuscript in preparation. Eylem, O., van Straten, A., de Wit, L., Rathod, S., Bhui, K., Kerkhof., A.J.F.M, (2020). Reducing Suicidal Ideation among Turkish Migrants in the Netherlands and in the UK: A Feasibility and Pilot RCT of an Online Intervention. Manuscript submitted for publication2 P Halvorsrud, K., Eylem, O., Bhui, K. (2020). Identifying evidence of effectiveness of photovoice: A systematic review and meta-analysis of the international healthcare literature. Manuscript in preparation. Van Bergen, D., Eylem, O., Montesinos A. (2019). Attempted suicide of Turkish young women in Europe and Turkey: A systematic literature review of socio-cultural, psychological, psychiatric and migration related factors. Manuscript submitted for publication2

2 Included in this thesis

a

acknoWledgements 280 ― Acknowledgements

ince my childhood, I have always been the curious type who wanted to see the world. Of all my journeys, my PhD has been the longest and the most challenging ones including many exciting and happy moments and also disappointing and frustrating ones. Coming to this point would not have Sbeen possible without the support, inspiration and faith of several people mentioned below and many more. Thus, this section expresses my appreciation to all of them for making this journey possible and meaningful with their contributions. First and foremost, I owe my deepest gratitude to my first supervisor Ad Kerkhof for providing me this opportunity. Ad, thank you for your trust and confidence in me. Throughout my PhD journey, you always encouraged me to think outside the box and expressed your faith in my passion and creativity. In the beginning, I didn`t have the scientific language to express my ideas. Thanks for being my sounding board. Having someone listening to me and offering feedback when needed helped me to grow professionally and also personally. I will always be grateful to you for that and will miss our conversations terribly. I would also like to thank my second supervisor Annemieke van Straten for her continued support and guidance throughout this journey. My academic writing has improved thanks to all the hours that Annemieke spent hoovering my manuscripts. Furthermore, I would like to express my warmest gratitude to the third supervisor of my PhD, Kamaldeep Bhui. Kamaldeep`s enthusiasm, professionalism and his innovative ideas are a great inspiration source to me. When times were tough, his guidance, exceptional kindness, and his invaluable help made my working environment a very good and a satisfactory one. I was honoured to be his PhD student and look forward to working with him in future. I am also very grateful to the co-supervisor of my PhD, Leonore de Witt. Leonore`s energy and immense passion always encouraged me. In particular, I would like to thank her for being always available to answer my questions, assist me in various ways and give me her insightful input. It has been a great pleasure to work with her. Next, I would like thank all my co-authors and would like to single out some special people who provided guidance and invaluable input in completion of this thesis. First, I Acknowledgements ― 281 would like to thank Pim Cuijpers. Needless to say, Pim`s outstanding work and expertise in systematic reviews and meta-analyses inspired my work and contributed to the quality of the meta-analysis in this thesis. I would like to express him my sincerest gratitude for giving me the opportunity to work closely with him in several projects, and I look forward to collaborating with him in future. Moreover, I also thank Diana van Bergen for giving me valuable insights on qualitative research and for working closely with me on the qualitative paper and the narrative systematic review which are included in this thesis. Specifically, I praise Diana for being open to ideas which I found unique in academia. She has been a great example for a researcher who is supporting her perspective based on a sound framework. I am also thankful for Burçin Ünlü İnce`s valuable support in several chapters of this PhD and her friendship from the very beginning of my journey. Further, I gratefully acknowledge Shanaya Rathod`s contribution. Her expertise in the cultural adaptation science influenced the key decisions about the cultural adaptation methodology in this thesis. I must also acknowledge the precious help of Marcello Gallucci, Koen Neijenhuijs, Aart Franken, and İlker Dalğar for supporting me in statistics and helping me to tackle challenging methodological issues. Further, I would like to show my appreciation to Mehmet Eskin, Ceren Açartürk and Meryam Scholer-Ocak. Their important achievements and expertise in their respective fields were a great inspiration source for me the rest of my career. I consider myself privileged for having their input on my manuscripts. I would also like to thank Erminia Colucci. Her A expertise in visual anthropology influenced my decisions whilst tackling the challenges with regards to engaging my population of interest in this thesis. I am very thankful for her support and her creative input in public engagement events that we did together. I am looking forward to working together with her in innovative public engagement projects in future. Furthermore, I would also like to express my appreciation to my beloved colleague and a friend Ümit Çetin. His work on Alevi-Kurdish communities and his bright ideas have opened new horizons for exciting projects in future. Finally, I would like to thank İnci Tebiş Picard and Zeynep İlkkurşun for their invaluable support 282 ― Acknowledgements in the implementation phase of my online intervention. Their perseverance and positivity certainly motivated me to keep on going despite the challenges of the recruitment. There are also some people whom I met in different stages of my student life. These people motivated me to pursue a PhD, and I would like to show my appreciation by mentioning them here. Here are some of my beloved teachers from my undergrad years: Müge Beidoğlu, Ali Civelek, Hüseyin Gürşan, Mehmet Çağlar, Ayten Sururi and Paula Fitzgibbon. Thank you all for pushing me to do better. Paula, I remember that one time you told me to push my brain and never get lazy. Of all the conversations we had, that sentence struck me the most, and even to this day it helps me to keep on going especially when I feel frustrated. Furthermore, I consider myself very lucky that I was supervised by Dennis Ougrin during my Masters in the Institute of Psychiatry, Kings College University of London. He is the reason why I developed interest in the suicidology field, and I do hope to work with him again in near future. Finally, Emel Stroop is another inspiring professional whom I would never forget. I was privileged to attend her CBT trainings, and I am very grateful for her support when I was applying for the scholarship to go to the Beck Institute. Further, I would like to express my appreciation to several organisations for opening their door for me, and supporting me throughout my PhD journey. First, I would like to show my sincere appreciation to the Beck Institute for Cognitive Behavior Therapy for selecting me as one of the student scholars out of over 400 applications worldwide back in 2013. I was truly honoured to meet prof. dr. Aaron T. Beck and Dr Judith Beck in person in Philadelphia in the US. Furthermore, I would like to acknowledge the European Commission for awarding me with a personal grant to conduct my PhD research. The opportunity that they gave me strengthened my belief in the European values such as equality of opportunities. Furthermore, I am very grateful to have worked with the Minddistrict in developing and implementing my online intervention. Specifically, I would like to thank Danielle Revers and Carlijn Bult for being very supportive, and welcoming throughout the process. Further, the Women`s Platform UK (WPUK) is another sentimental place which makes me feel home. In Acknowledgements ― 283 particular, I would like to express my warmest gratitude to Nilgün Yıldırım for being my mentor and her creative input in the implementation phase of my online intervention. Our working relationship has been a good example for a partnership between the third sector organisations and academia, thereby our work was awarded with a few community engagement awards. I would also like to thank the European Federation of Psychology Students Association (EFPSA). Through them I had the opportunity to work with some bright and wonderful ladies: Lena Steubl, Gözde Topgüloğlu Danışman, Zaneta Eleni Melissourgaki, Paula Stroian, Daria Sfeci, Anna Kalushe and Başak Özkara. They voluntarily supported the implementation of my online intervention and contributed to the meta-analysis that was included in this thesis. Moreover, I would like to acknowledge the Centre for Turkey Studies (CEFTUS) for working together with me in organising a public forum at the House of Parliament in London in the UK. Undoubtedly, their involvement contributed to the awareness raising about the issues around suicide in Turkish, Kurdish and Turkish Cypriot diaspora not only on an individual level but also on societal and on political levels. I would also like to thank the Centre for Public Engagement, Emerald Publishing Group, and the Centre for Mind in Society Queen Mary University of London for recognising and supporting my public engagement work. Finally, I would like to extend my gratitude to Derman for not only supporting me during my PhD but also for giving me an opportunity to work with them in the suicide awareness project concerning a partnership between Derman and the East London A NHS Foundation Trust. I must acknowledge that working with them improved my knowledge on real world implications of psychological interventions and contributed to my flexibility as a field researcher. It goes without saying that I also happened to meet some special people through them. Here are some I singled out: Aradhana Perry, Angela Byrne, Ali Aksoy, Nurullah Turan, Ufuk Genç and Nursel Taş. I would like to thank all my colleagues at the department of Clinical Psychology, VU Amsterdam University and at the Centre for Psychiatry, Queen Mary University of London. I am very grateful that our paths have crossed with so many talented and inspirational colleagues. Especially, I would like to show my gratitude to my paranymphs 284 ― Acknowledgements

Anne de Graaff and Yağmur Amanvermez who gave me enormous support for organising my defence. Their involvement made the process fun, and their enthusiasm and support were invaluable. Moreover, I would like to thank Baran Metin (also known as Mehmet) for his involvement in my social media campaigns. Even though that didn`t bring him fame, he certainly conquered some hearts with his acting talents. I am also thankful to have met Kristoffer Halvorsrud, and I look forward to continue working with him in exciting photovoice projects. Furthermore, I would like to thank those colleagues for the great memories; Dalya Samur, Livia Sani, Matthijs de Wit, Karen Holtmaat, Ellen Driessen, Robin Kok, Edit van`t Hof, Derek de Beurs, Wouter Ballegooijen, Mirjam Reijnders, Nicolaos Boumparis, Claudia Buntrock, Patricia Otero, Dilfa Juniar, Metta Rahmadiana, Ferhat Jack İçöz, Antoine Selim Bilgin, and Jeyda İbrahim Özlü. Moreover, I would like to acknowledge the precious help that I had from Lisa Kass, Renata Samulnik, Sherida Slijngaard, and Johan Meester in various administrative tasks. Thank you for your patience and understanding. Finally, I would like to extend my gratitude to Marco Bernard and Ralp de Vries for their expertise in their fields and for being very nice to work with. I would like to thank all the members of my PhD thesis committee (Dr. Derek de Beurs, Dr. Fatima El Fakiri, prof. dr. Joop de Jong, prof. dr. Marrie Bekker, prof. dr. Karien Stronks and prof. dr. Marcus Huijbers) for giving their valuable time and expertise in evaluating the present work. Sıra geldi sevgili Annem, Babam ve Kardeşim`e. Her zaman benim potansiyelime inandığınız, hep daha iyisini yapmama teşvik ettiğiniz ve bana imkanlar sunduğunuz için çok teşekkür ederim. Özellikle Annem ve Babam`ın yaptığı fedakarlıkları ve beni daha ileriye iten motivasyonu olmasa ben bu noktaya gelemezdim. Bu doktora sizinde başarınızdır. Last but not least, I would like to thank my Dutch family (van Bergeijks) and my partner, the love of my life, Michiel Pieter van Bergeijk for making this country my second home. Undoubtedly, Michiel has seen all my ups and downs throughout my PhD journey. He listened to everything, and always managed to calm me down by telling Acknowledgements ― 285 me “sit and think” during my temperemental Mediterranean moments. Even in the toughest times, he always had confidence in me, and always assured me that the finish line was getting closer day by day. Thank you for not only being my partner but also for being my best friend. Thank you for being there for me bulli.

A

r

references 288 ― References

Adam, K.S., Sheldon-Keller, A.E., West, M. (1996). Attachment organisation and history of suicidal behavior in clinical adolescents. Journal of Consulting and Clinical Psychology, 64, 264-272. Adewuya, AO., Makanjuola Roger, OA. (2008). Social distance towards people with mental illness in southwestern Nigeria. Australian and New Zealand Journal of Psychiatry 42, 389-395. Ahn, Ji-Hoon, Kim, Won-Hyoung, Choi, Hye-Jin, Jeon, Jin-Yong, Song, In-Gyu, Jae-Nam Bae. (2015). Stigma of Mental Illnesses as Perceived by North Korean Defectors Living in South Korea. Korean Neuropsychiatric Association,12,9-15. Aichberger MC, Heredia Montesinos A, Bromand Z, Yesil R, Temur-Erman S, Rapp MA, Heinz A, Schouler-Ocak M. (2015). Suicide attempt rates and intervention effects in women of Turkish origin in Berlin. Eur Psychiatry,30,480-5. Aichberger, M.C., Bromand, Z., Rapp, M.A., Rapp, M.A., Yesil, R., Heredia Montesinos, A. et al., (2015a). Perceived ethnic discrimination, acculturation, and psychological distress in women of Turkish origin in Germany. Soc Psychiatry Psychiatr Epidemiol,50,1691–1700. Aichberger, M.C., Yesil, R., Rapp, M.A., Schlattmann, P., Temur-Erman, S., Bromand, Z. et al., (2013) Surveying migrant populations – methodological considerations: An example from Germany, International Journal of Culture and Mental Health, 6:2, 81-95. Akın, D., Tuzun, Y.,Cil, T. (2017). Türkiye’nin Güneydoğusundaki Akut Zehirlenme Olgularının Profili [The Profile of Acute Poisonings in South East of Turkey] Dicle Tıp Dergisi.3,195-198. Akpınar, A. (2003). The honour/ shame complex revisited: violence against women in the migration context. Women’s Studies International Forum,26,425-442. Aliverdinia, A., Pridemore, W.A. (2009). Women’s fatalistic suicide in Iran: a partial test of Durkheim in an Islamic Republic. Violence against women.15,307-320. Altindag, A., Mustafa, O., Remzi, O. (2005). Suicide in Batman, South eastern Turkey. Suicide and Life Threatening Behavior.35,478-482. Anglin, D.M., Link, B.C., Phelan, J.C. (2006). Racial differences in stigmatizing attitudes toward people with mental illness. Psychiatric Services 6, 857-862. Arshad, U., Ul-Ain, F., Gauntlett, J., Nusrat, H., Chaudhry, N., Taylor, P. (2019). A systematic review of the evidence supporting mobile and internet-based psychological interventions for self-harm. Suicide Life Threat Behav; https://doi.org/10.1111/sltb.12583. Asirdizer, M., Yavuz, S., Demirag, A., Dizdar, G. M. (2010). Suicides in Turkey Between 1996 and 2005 General Perspective. Am J Forensic Med Pathol.31,138–14. Atlı Z., Eskin M., Dereboy C. (2009). The validity and the reliability of suicide probability scale (SPS) in clinical sample. Klinik Psikiyatri,12,111-24. Augsberger, A., Yeung, A., Dougher, M., Hahm, H.Y. (2015). Factors influencing the underutilization of mental health services among Asian American women with a history of depression and suicide. BMC Health Services Research 15:542. http://dx.doi.org/10.1186/s12913-015-1191. Aydemir, O., Guvenir, T., Kuey, L., Kultur, S. (1997). The validity and reliability of the Turkish version of Hospital anxiety and depression scale [Article in Turkish]. Türk Psikiyatri Dergisi, 8,280-287. Aznar-Lou, I., Serano-Blanco, A., Fernandez, A., Luciano, J.V., Rubio-Valera, M. (2016). Attitudes and intended behaviour to mental disorders and associated factors in Catalan population, Spain: cross- sectional population-based survey BMC Public Health 16:127, https://doi.org/10.1186/s12889-016- 2815-5. Babtiste, V.M. (2017). The Impact of Stigma on the Mental Health of Resettled African and Asian Refugees. Graduate College Dissertations and Theses. 785. https://scholarworks.uvm.edu/ graddis/785 References ― 289

Bağlı, M., Sever, A. (2003). Female and male suicides in Batman, Turkey: poverty, social change, patriarchal oppression and gender links.The Women’s Health & Urban Life: An International and Interdisciplinary Journal, 2, 60-84. Baker, A. E. Z., Procter, N. G. (2016). Engaging with culturally and linguistically diverse communities to reduce the impact of depression and anxiety: a narrative review; Health and Social Care in the Community, 24, 386–398. Bakker, W., Van Oudenhoven, J. P., van der Zee, K. I. (2004). Attachment styles, personality, and Dutch emigrants’ intercultural adjustment. European Journal of Personality, 18, 387–404. Bangor, A., Kortum, P.T., Miller, J.T. (2008) An empirical evaluation of the system usability scale. International Journal of Human-Computer Interaction, 24,6, 574-594. https://doi. org/10.1080/10447310802205776 Bartholomew, K., Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four- category model. Journal of Personality and Social Psychology, 61, 226–244. Bateman, A., Fonagy, P. (1999). Effectiveness of Partial Hospitalization in the Treatment of Borderline Personality Disorder: A Randomized Controlled Trial. Am J Psychiatry,156,1563–1569. Bateman, A., Fonagy, P. (2008). 8-Year Follow-Up of Patients Treated for Borderline Personality Disorder: Mentalization-Based Treatment Versus Treatment as Usual. Am J Psychiatry, 165, 631–638. Baumeister, R.F. (1990). Suicide as escape from self. Psychological Review, 97, 90–113. Baysu, G., Phalet, K., Brown, R. (2013). Relative group size and minority school success: The role of intergroup friendship and discrimination experiences. British Journal of Social Psychology,53, 328-349. Beautrais, A. L. (2006). Women and suicidal behaviour. Crisis,27,153-156. Beck, J.S. (1995). Cognitive therapy: Basics and beyond. New York: Guilford. Beck A.T., Steer R.A., Ranieri W.F. (1988). Scale for suicide ideation: Psychometric properties of a self- report version. J ClinPsychol, 44,499-505. Beck A.T. (2005) The current state of cognitive therapy. Arch Gen Psychiatry 62, 953–9. Beck, A.T., Steer, R.A. (1988). Manual for the Beck Hopelessness Scale. San Antonio, TX: Psychological Corp. Beck, A. T., Weissman, A., Lester, D., Trexler, L. (1974). The measurement of pessimism: The Hopelessness Scale. Journal of Consulting & Clinical Psychology, 42, 861–965. Beck, A.T., Steer, R.A. (1993). Beck Scale for Suicidal Ideation Manual (BSS).The Psychological Corporation. Harcourt Brace & Company Beck, A.T., Kovacs, M., Weissman, A. (1979). Assessment of suicidal intention: the scale for suicide R intention. Journal of Consulting and Clinical Psychology,47,343-352. Beck, A.T., Schuyler, D., Herman, I. (1974). The Prediction of Suicide. Oxford, England: Charless Press Publishers. Beck, A.T., Ward, C. H., Mendelson, M., Mock, J., Erbaugh, J. (1961). An inventory for measuring depression. Archieves of General Psychiatry,561-572. Begum S (2006). Bangladeshi Stop Tobacco Project. Quarterly Monitoring Reports London: Tower Hamlets Bangladeshi Stop Tobacco Project. Behrens, K., del Pozo, M.A., Grobhenning, A., Sieber, M., Graef-Calliess, I.T. (2015). How much orientation towards the host culture is healthy? Acculturation style as risk enhancement for depressive symptoms in immigrants. International Journal of Social Psychiatry, 61, 498 –505. Belone, L., …Wallerstein, N. (2016). Community-Based Participatory Research Conceptual Model: Community Partner Consultation and Face Validity. Qualitative Health Research, 26, 117–135. 290 ― References

Bernal, G., J. Bonilla, C. Bellido. (1995). “Ecological validity and cultural sensitivity for outcome research: issues for the cultural adaptation and development of psychosocial treatments with Hispanics.” J Abnorm Child Psychol 23 (1):67-82. Berry, J. W. (2009), A critique of acculturation. International Journal of Intercultural Relations, 33, 361–371. Berry, J. W. (1997). Immigration, acculturation, and adaptation. Applied Psychology: An International Review, 46(1), 5–68. Berry, J. W., Sabatier, C. (2011). Variations in the assessment of acculturation attitudes: Their relationship with psychological wellbeing. International Journal of Intercultural Relations, 35, 658-669 Bertolote, J.M., Fleischmann, A. (2005). Suicidal behavior prevention: WHO perspectives on research. American Journal of Medical Genetics. https://doi.org/10.1002/ajmg.c.30041 Bertolote, J.M., Fleischmann, A. (2009). A global perspective on the magnitude of suicide mortality. In D. Wasserman & C. Wasserman (Eds.), Suicidology and suicide prevention. A global perspective (pp. 91-98). Oxford, UK: Oxford University Press. Bevington D., Fuggle, P., Fonagy, P. (2015). Applying attachment theory to effective practice with hard- to-reach youth: the AMBIT approach, Attachment & Human Development, 17:2, 157-174. http:// dx.doi.org/10.1080/14616734.2015.1006385 Bharadwaj, P., Pai, M.M., Suziedelyte. (2017). Mental health stigma. Economics Letter 159,57-60. Bhugra , D. (2004). Migration and mental health. Acta Psychiatrica Scandinavica, 109, 243 – 258 . Bhugra, D.,Gupta, S. (2011). Migration and mental health. Cambridge MA: Cambridge University Press. Bhugra, D. (2002). Suicidal behaviour in South Asians in the UK. Crisis,23, 108-113. Bhugra, D., Baldwin, S., Desai, M. & Jacob, K.S. (1999a). Attempted suicide in West London II. Inter group comparisons. Psychological Medicine,29,1131-1139. Bhugra, D., Bhui, K., Desai, M., Singh, J., Baldwin, D. (1999b). The Asian cultural identity schedule: an investigation of culture and deliberate self-harm. International Journal of Methods in Psychiatry Research,8,212-218. Bhugra, D., Desai, M., Baldwin, D.S. (1999c). Attempted suicide in west London, I. Rates across ethnic communities. Psychological Medicine,29,1125-1130. Bhui, K.S., Lenguerrand, E. , Maynard , M.J., Stansfeld, S.A., Harding, S. (2012). Does cultural integration explain a mental health advantage for adolescents? International Journal of Epidemiology, 41, 791 – 802 . Bhui, K. (2008). Migration and Mental Health pp. 184-209 in Freeman, H., Stansfeld, S., (2008). The impact of the environment on psychiatric disorder. Publisher: Routledge/Taylor & Francis Group. Bhui, K. (2010a). Commentary: Religious, cultural and social influences on suicidal behaviour. International Journal of Epidemiology,39,1495-1496. Bhui, K. (2010b). Culture and complex interventions: Lessons for evidence, policy and practice. British Journal of Psychiatry 197: 172-173, https://doi.org/10.1192/bjp.bp.110.082719 Bhui, K. (2011). Cultural psychiatry and epidemiology: Researching the means, methods and meanings. Transcultural Psychiatry,48,90-103. Bhui, K. (2019). Social determinants of mental health. In: Bhugra, D., Bhui, K., Yeung Shan Wong, S., Gillman, S.E. Oxford Textbook of Public Mental Health p. 173-177 Bhui, K., Halvorsrud, K., Nazroo, J. (2018). Making a difference: ethnic inequality and severe mental illness. The British Journal of Psychiatry, 213, 574–578. doi: 10.1192/bjp.2018.148. Bhui, K., McKenzie, K., Rasul, F. (2007). Rates, risk factors & methods of self-harm among minority ethnic groups in the UK: A systematic review. BMC Public Health, 16, 145-151. References ― 291

Bhui, K., Stansfeld, S., Hull, S., Priebe, S., Mole, F., Feder, G. (2003). Ethnic variations in pathways to use of specialist mental health services in the UK. British Journal of Psychiatry, 182,105-116. Bhui, K.S., Lenguerrand, E., Maynard, M.J., Stansfeld, S.A., Harding, S. (2012). Does cultural integration explain a mental health advantage for adolescents? Int J Epidemiol,41, 791-802. Binbay T., Direk N., Aker T., Akvardar Y., Alptekin K., Cimilli C. (2014). Psychiatric epidemiology in Turkey: main advances in recent studies and future directions. Turk Psikiyatri Derg, 2, 264-281. Bjelland I., Dahl A.A, Haug T.T., Neckelmann D. (2002). The validity of the Hospital Anxiety and Depression Scale. An updated literature review. J Psychosom Res, 52,69-77. Borges, G., Breslau, J., Su, M., Miller, M., Medina-Mora, M.E.,Aguilar-Gaxiola, S. (2009) ‘Immigration and Suicidal Behavior Among Mexicans and Mexican Americans’, American Journal of Public Health,99, 728–33. Bowlby, J. (1973). Attachment and loss. Vol. 2. Separation. New York, Basic Books. Bryant, R. A., Schafer, A., Dawson, K. S., Anjuri, D., Mulili, C., Ndogoni, L., van Ommeren, M. (2017). Effectiveness of a brief behavioural intervention on psychological distress among women with a history of gender-based violence in urban Kenya: randomized clinical trial. PLoS Medicine, 14, e1002371. Braun, V. Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology,3, 77-101. Brenner, M. H. (2019). Social class and mental health: The impact of international recessionand austerity. In: Bhugra, D., Bhui, K., Yeung Shan Wong, S., Gillman, S.E. Oxford Textbook of Public Mental Health. p.p. 159-173. Brown, C., Conner, K. O., Copeland, V. C., Grote, N., Beach, S., Battista, D., (2010). Depression stigma, race, and treatment seeking behavior and attitudes. Journal of Community Psychology, 38, 350–368. Brown, G. K. (2001). A review of suicide assessment measures for intervention research with adults and older adults. Retrieved on 21st September 2014, from http://sbisrvntweb.uqac.ca/archivage/15290520.pdf Brown, R. and Zagefka, H. (2011). The dynamics of acculturation: an intergroup perspective. Advances in Experimental Psychology,44,129-176. Bruaffers, R., Demytenaere, K., Hwang, I., Ciu, W.T., Sampson, N., Kessler, R.C. et al. (2011). Treatment of suicidal people around the world. British Journal of Psychiatry, 199, 64-70. Brücker, B., Muheim, F., Berger, P., Riecher-Rossler, A. (2011). Charakteristika von Suizidversuchen turkischer Migranten in Kanton Basel-Stadt. Resultate der WHO/EURO-Multizenterstudie [Characteristics of suicide attempts of Turkish migrants in the canton of Basel-Stadt-Results of the R WHO/EURO multicentre study] Nervelheilkunde,30,517-522. Burger, I., van Hemert, A. M., Bindraban, C. A., Schudel, W. J. (2006).Parasuicide in The Hague. Incidences in theyears 2000–2004 [Article in Dutch]. Epidemiologisch Bulletin, 40,2–8. Bursztein Lipsicas, C. Makinen, I.H., Wasserman, D., Apter, A., Kerkhof, A., Michel, K. et al. (2012). Gender distribution of suicide attempts among migrant groups in European countries-an international perspective. European Journal of Public Health. Advance online publication. https://doi. org/10.1093/eurpub/cks029 Çakmak, H. K. Altuntas, N. (2009). Reconsidering gender inequality and honor suicide within the frame of different liberal theories: Turkey-Batman case. Muslim World Journal of Human Rights,5,1554-4419. Canetto, S. S. (2015). Suicidal Behaviors Among Muslim Women Patterns, Pathways, Meanings, and Prevention. Crisis, 36, :447–458. Canetto, S.S. (2008). Women and suicidal behavior: a cultural analysis. American Journal of Orthopsychiatry,78,259-266. 292 ― References

Caplan, S. (2016). A Pilot Study of a Novel Method of Measuring Stigma about Depression Developed for Latinos in the Faith-Based Setting. Community Ment Health J, 52:701–709. Caplan, S., Paris, M., Whittemore, R., Desai, M., Dixon, J., Alvidrez, J., Escobar, J., Scahill, S. (2011). Correlates of religious, supernatural and psychosocial causal beliefs about depression among Latino immigrants in primary care, Mental Health, Religion & Culture, 14:6, 589-611, https://doi. org/10.1080/13674676.2010.497810 Caplan, S., Sosa Lovera, A. and Liberato, P.R. (2018). A feasibility study of a mental health mobile app in the Dominican Republic: The untold story. International Journal of Mental Health, 47, 311–345 https://doi.org/10.1080/00207411.2018.1553486 Castles, S. (2010). Understanding global migration: A social transformation perspective. Journal of Ethnic and Migration Studies,36, 1565-1586. CBS, Centraal Bureau voor de Statistiek (2016). Retrieved June 3rd 2016 from: http://statline.cbs.nl/ StatWeb/publication/?VW=T&DM=SLNL&PA=37296ned&D1=a&D2=0,10,20,30,40,50,(l-1)- l&HD=100428 1217&HDR=G1&STB=T. Cetin, F. C. (2001). Suicide Attempts and Self-Image Among Turkish Adolescent. Journal of Youth and Adolescence,30,641-651. Cetin, U. (2015). Durkheim, ethnography and suicide: Researching young male suicide in the transnational London Alevi-Kurdish community. Ethnography, 0, 1–28. Cetin, U. (2017). Cosmopolitanism and the relevance of ‘zombie concepts’: The case of anomic suicide amongst Alevi Kurd youth. The British Journal of Sociology,68,145-166. Chen, E. Y., Chan, W. S., Wong, P. W., Chan, S. S., Chan, C. L., Law, Y. W,....,Yip, P. S. (2006). Suicide in Hong Kong: a case-control psychological autopsy study. Psychological medicine,36, 815-825. Cheng, Z. H. (2015). Asian Americans and European Americans’ stigma levels in response to biological and social explanations of depression. Soc Psychiatry Psychiatr Epidemiol 50:767–776, https://doi. org/10.1007/s00127-014-0999-5 Christensen H., Farrer L., Batterham P. J., Mackinnon A., Griffiths K. M., et al. (2013) The effect of a web- based depression intervention on suicide ideation: secondary outcome from a randomised controlled trial in a helpline. BMJ Open 3: e002886. Christensen, H., Batterham, P. J., O’Dea, B. (2014). E-health interventions for suicide prevention. International journal of environmental research and public health, 11, 8193-8212. https://doi. org/10.3390/ijerph110808193 Chu, J. P., Huynh, L., Patricia, A. (2012). Cultural adaptation of evidence-based practice utilizing an iterative stakeholder process and theoretical framework: problem solving therapy for Chinese older adults. Int J Geriatr Psychiatry, 27: 97–106. Cihangir, S. (2013). Gender specific honor codes and cultural change. Group Processes and Interpersonal Relations,16, 319-333. Clarke, J., Proudfoot, J., Whitton, A., Birch, M.-R., Boyd, M., Parker, G., . . . Fogarty, A. (2016). Therapeutic alliance with a fully automated mobile phone and web-based intervention: Secondary analysis of a randomized controlled trial. JMIR mental health, 3, e10-e10. https://doi.org/10.2196/ mental.4656 Clemens, C. A. L. (2011). Suicide girls: Orhan Pamuk’s snow & the politics of resistance in contemporary Turkey. Feminist Formations,23, 138-154. Clement, S., Schauman, O., Graham, T., Maggioni, F., Evans-Lacko, S., Bezborodovs, N., Morgan, C., Rüsch N, Browwn, JSL., Thornicroft G. (2015a). What is the impact of mental health-related stigma on help-seeking? A systematic review of quantitative and qualitative studies. Psychological Medicine 1-17. http://dx.doi.org/10.1017/S0033291714000129. References ― 293

Clement, S., Williams, P., Farelly, P., Hatch, SL., Schauman, O., Jeffery, D., Henderson, C., Thornicroft, G. (2015b). Mental Health–Related Discrimination as a Predictor of Low Engagement With Mental Health Services. Psychiatric Services 66, 171-176. Cohen J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Erlbaum, Hillsdale, NJ. Colucci, E. (2013). Culture, cultural meaning(s), and suicide. In D. Lester, E. Colucci (Eds.), Suicide and culture: Understanding the context. Gottingen, Germany: Hogrefe Publishers. Colucci, E., Heredia-Montesinos, A. (2013). Violence against women and suicide in the context of migration: A review of the literature and a call for action. Suicidology Online, 4,81-91. Colucci, E., San Too, L. (2015). Culture, cultural meanings, and suicide among people from migrant and refugee backgrounds In van Bergen, D., Montesinos, A.H. & Schouler-Ocak, M. (2015). Suicidal behavior of immigrants and ethnic minorities in Europe. Hogrefe Publishers. Conelly, R., Gayle, V., Lambert, P.S. (2016). Ethnicity and ethnic group measures in social survey research. Method Innov, 9, 1-10. Conner, K.R., Beautrais, A. L., Brent, D. A., Conwell., Y., Phillips, M. R., Schneider, B. (2012). The Next Generation of Psychological Autopsy Studies Part 2. Interview Procedures. Suicide and Life- Threatening Behavior, 42, 86-103. Consort Statement Checklist - Consolidated Standards Of Reporting Trials (CONSORT). (http://www. consort-statement.org/). Accessed 31 may 2018. Cooper, C., Spiers, N., Livingston, G., Jenkins, R., Meltzer, H., Brugha, T., McManus, S., Weich, S., Bebbington, P. (2013). Ethnic inequalities in the use of health services for common mental disorders in England. Soc Psychiatry Psychiatr Epidemiol,48,685–692. Cooper, J., Murphy, E., Webb, R., Hawton, K., Bergen, H., Waters, K., Kapur, N. (2010). Ethnic differences in self-harm, rates, characteristics and service provision: three-city cohort study. The British Journal of Psychiatry,197, 212–218. Copelj, A., Kiropoulos, L. (2011). Knowledge of depression and depression related stigma in immigrants from former Yugoslavia. J Immigrant Minority Health 13,1013–1018. https://doi.org/10.1007/s10903- 011-9463-8. Corrigan P, Rüsch N. (2002). Mental illness stereotypes and clinical care: do people avoid treatment because of stigma? Psychiatric Rehabilitation Skills 6, 312–334. Corrigan, P. W. (2016). Lessons learned from unintended consequences about erasing the stigma of mental illness. World J Psychiatry, 15,67-73. R Corrigan P. W. (2004). How stigma interferes with mental health care. American Psychologist 59, 614–625. Corrigan P. W., Watson A. C. (2007). The stigma of psychiatric disorders and the gender, ethnicity, and education of the perceiver. Community Ment Health J, 43, 439–58. Corrigan, P. W., Watson, A. C. (2002). Understanding the impact of stigma on people with mental illness. World Psychiatry,1,16-20. Craig P, Dieppe P, Macintyre S, Michie S, Nazareth I, Petticrew M; Medical Research Council Guidance (2008). Developing and evaluating complex interventions: the new Medical Research Council guidance. British Medical Journal 337, a1655 Crenshaw, K., (1989). Demarginalizing the intersection of race and sex: a black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. Univ. Chic. Leg. Forum 139, 139–167. Cuijpers, P. (2016). Calculating and pooling effect sizes. In: Meta-analyses in Mental Health Research: A Practical Guide. Vrije Universiteit Amsterdam, The Netherlands. p. 67-91. 294 ― References

Cuijpers, P., Cristea, I. (2015). What if a placebo effect explained all the activity of depression treatments? World Psychiatry, 14:3. Cuijpers P, Munoz RF, Clarke GN, Lewinsohn PM. (2009). Psychoeducational treatment and prevention of depression: the “Coping with Depression” course thirty years later. Clinical Psychology Review 29, 449-458. Cuijpers, P., Donkers, T., van Straten, A., Li, J., Andersson, G. (2010). Is guided self-help as effective as face- to-face psychotherapy for depression and anxiety disorders? A systematic review and meta-analysis of comparative outcome studies. Psychological Medicine, 40, 1943-1957 Cuijpers, P., Eylem, O., Karyotaki, E., Zhou, X., & Sijbrandij, M. (2019). Psychotherapy for depression and anxiety in low- and middle-income countries. In D. Stein, & J. Bass (Eds.), Global Mental Health and Psychotherapy: Adapting Psychotherapy for Low- and Middle-Income Countries (pp. 173-192). Elsevier. https://doi.org/10.1016/b978-0-12-814932-4.00008-2Cuijpers, P., Kleiboer, A., Karyotaki, E., Riper, H. (2017). Internet and mobile interventions for depression: Opportunities and challenges. Depression and Anxiety. https://doi.org/10.1002/da.226-41 Cull, J. G., Gill, W. S. (1990). Suicide probability scale Manual. Los Angeles, Western Psychological Services. D`angelo, A., Galip, O., Kaye, N. (2013). Welfare Needs of Turkish and Kurdish Communities in London. The Social Policy Research Centre, London, UK. De Beurs, D., Kirtley, O., Kerkhof, A., Portzky, G., O Connor, R. (2014). The Role of Mobile Phone Technology in Understanding and Preventing Suicidal Behavior, Crisis, 36,79–82. https://doi. org/10.1027/0227-5910/a000316. De Jaegere, E. , van Landschoot, R., van Heeringen, K., van Spijker, B.A.J., Kerkhof, A.J.M.F., Mokkenstorm, J.K., Portzky, G. (2019). The online treatment of suicidal ideation: a randomised controlled trial of an unguided web-based intervention. Bahaviour Research and Therapy,119, https://doi.org/10.1016/j.brat.2019.05.003. De Jong, M. L. (1992). Attachment, individuation, and risk of suicide in late adolescence. Journal of Youth and Adolescence, 21 (3), 357-373. De Leo, D., Burgis, S., Bertolote, J.M., Kerkhof, A., J., F., M., Bille-Brahe, U. (2006). Definitions of suicidal behaviour lessons learned from the WHO/EURO multicentre study. Crisis, 27, 4-15. Department of Health (2012). Preventing suicide in England: A cross-government outcomes strategy to save lives. Assessment of impact on equalities.: Department of Health. Department of Health. (2017). Preventing suicide in England: Third progress report of the cross- government outcomes strategy to save lives.: Department of Health. De Pater, I. E., Van Vianen, A. E. M., Derksen, M. (2003). Het aangaan van hechte relaties en de Cross culturele aanpassing van expatriates: de rol van persoonlijkheid en hechtingsstijl. Gedrag in Organisaties, 16, 89–107. Devries, K. M., Mak, J. Y., Bacchus, L. J., Child, J. C., Falder, G., Petzold, M., Astbury, J., Watts, C. H. (2013). Intimate Partner Violence and Incident Depressive Symptoms and Suicide Attempts: A Systematic Review of Longitudinal Studies. Plos Medicine.10:5. Devrimci-Ozguven H., Sayil I. (2003). Suicide attempts in Turkey: results of the WHO-EURO Multicentre Study on Suicidal Behavior. Can J Psychiatry, 48, 324-329. Diehl, C., Koening, M., Ruckdeschel, K. (2009). Reigiosity and gender equality: comparing natives and Muslim migrants in Germany. Ethnic and Racial Studies, 32,278-301. Dilbaz, N., Bitlis, V., Bayam, G., Berksun, O., Holat, H., Tuzer, T. et al. (1995). Suicide intent scale: validity and reliability [Article in Turkish]. Psychology, Psychiatry and Psychopharmacology, 3, 28-31. References ― 295

Douki, S., Zineb, S. B., Nacef, F., Halbreich, U. (2007). Women’s mental health in the Muslim world: Cultural, religious, and social issues. Journal of Affective Disorders, 102, 177–189. Dueweke, A. R., Hurtado, G., Hoevey, J. D. (2015). Protective psychosocial resources in the lives of Latina migrant farmworkers. Journal of Rural Mental Health,39,162-177. Dumensnil, H., Verger, P. (2009). Public awareness campaigns about depression and suicide: A review. Psychiatric Services, 60,1203-1213. Durant, T., Mercy, J., Kresnow, M., Simon, T., Potter, L., Hammond, W. (2006). Racial differences in hopelessness as a risk factor for a nearly lethal suicide attempt. Journal of Black Psychology, 32, 285–302. Duval S., Tweedie R. (2000). Trim and fill: A simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 56, 455–463. Durkheim E. (1952) [1887]Suicide. A Study in Sociology. London, Penguin Books. Eisenberg, D., Downs, MF., Golberstein, E., Zivin, K. (2009). Stigma and help seeking for mental health among college students. Medical Care Research and Review 1-20, https://doi. org/10.1177/1077558709335173 Eldering, L. (2014). Cultuur en Opvoeding. [Culture and Education]. Rotterdam, Leminiscaat. Enginyurt, O., Ozer, E., Gumus, B., Demiz, E. Y., Cankaya, S. (2014). Evaluation of suicide cases in Turkey, 2007–2012. Med Science Monitor,20, 614-62. Enneli, P., Modood, T., Bradley, H. (2005). Young Turks and Kurds: A Set of “invisible” Disadvantaged Groups. Joseph Rowntree Foundation, London, UK. Eroğlu M, Yıldırım A, Uzkeser M, Saritas A, Acemoglu H, Navruz M, Emet M. (2014). Emergency room visits for suicide attempts: rates, trends and sociodemographic characteristics of suicide attempts in Northeastern Anatolia. Klin. Psikofarmakol. Bül,24, 350-359. Ersan, E. E., Kilic, C. (2013). Sivas Numune Hastanesi Acil Servisine Basvuran Intihar Girisimlerinin Degerlendirilmesi [Evaluation of Suicide Attempts Referring to Sivas Numune Hospital Emergency Department] Klinik Psikiyatri,16,98-109. Ersanilli, E., Koopmans, R. (2010). Rewarding integration? Citizenship regulations and the socio- cultural integration of immigrants in the Netherlands, France and Germany. Journal of Ethnic and Migration Studies,36,773-791. Ersanilli, E. (2010). Comparing Integration: Host Culture Adoption and Ethnic Retention Among Turkish Immigrants and Their Descendants in France, Germany and the Netherlands. Amsterdam, Unpublished PhD thesis, VU University Amsterdam. R Eser E., Dinç G., Cambaz S. EURO-QoL (EQ-5D) indeksinin toplum standartları ve psikometrik özellikleri: Manisa kent toplumu örneklemi. 2.Sağlıkta Yaşam Kalitesi Kongresi: 5-7 April 2007; Izmir Izmir: Bildiri Özetleri Kitabı Meta Basımevi; 2007, 78. Eskin, M. (1993). Reliability of the Turkish version of the perceived social support from friends and family scales, scale for interpersonal behavior, and suicide probability scale. Journal of clinical psychology, 49 (4), 515-522. Eskin, M. (1999a). Gender and cultural differences in the 12 month prevalence of suicidal thoughts and attempts in Swedish and Turkish adolescents. Journal of Gender, Culture and Health, 3, 87-200. Eskin, M. (1999b). Social reactions of Swedish and Turkish adolescents to a close friends` suicidal disclosure. Social Psychiatry Psychiatr Epidemiol,34,492-497. Eskin, M. (2003). A cross cultural investigation of the communication of suicidal intent in Swedish and Turkish adolescents. Scandinavian Journal of Psychology,44,1-6. Eskin, M. (2004). The effects of religious versus secular education on suicide ideation and suicidal attitudes in adolescents in Turkey. Social Psychiatry and Psychiatr Epidemiol, 39,536-542. 296 ― References

Eskin, M. (2017). Gencler arasinda intihar dusuncesi, girisimi ve tutumlari [Attitudes towards suicide, ideation and attempts among young people]. Turk Psikoloji Dergisi, 32(80), 93–111. Eskin, M., Palova, E., Krokavcova, M. (2014). Suicidal behaviour and attitudes in Slovak and Turkish high school students: A cross-cultural investigation. Archieves of Suicide Research,18,58-73. Eskin, M., Poyrazli, S., Janghorbani, M., Bakhshi, S., Carta, M. G., Moro, M. F., … Taifour, S.(2019). The Role of Religion in Suicidal Behavior, Attitudes and Psychological Distress Among University Students: A Multinational Study. Transcultural Psychiatry, https://doi.org/10.1177/1363461518823933 Eskin, M., Schild, A., Oncu, B., Stieger, S., Voracek, M. (2015). A cross-cultural investigation of suicidal disclosures and attitudes in Austrian and Turkish university students. Death Studies,39,584-591. Eylem, O., Dalgar, I., Tok, F., Ünlü İnce, B., van Straten, A., Kerkhof, A.J.F.M., Bhui, K. (2019). The association between acculturation and suicidal ideation among Turkish migrants in the Netherlands. Psychiatry Research,275,71-77. Eylem, O., van Straten A., Bhui K., Kerkhof A.J.F.M. (2015). Protocol: Reducing suicidal ideation among Turkish migrants in the Netherlands and in the UK: Effectiveness of an online intervention. Int Rev Psychiatry, 27: 72–81. Eylem, O. (2011). Development and Preliminary Validation of the Self-Harm Questionnaire. LAP. Lambert Academic Publishing. Eylem, O., de Wit, L., van Straten, A. et al. (2020) Stigma for common mental disorders in racial minorities and majorities a systematic review and meta-analysis. BMC Public Health 20, 879 . https://doi. org/10.1186/s12889-020-08964-3Eylem, O., Eskin, M. (2020). Suicidal Behaviour and Social Cultural Factors Among Immigrants of Turkish Origin in Germany. In Scholer-Ocak, M., Khan, Moosa M. Suicide Across Cultures. Oxford Cultural Psychiatry Series, UK. Eylem, O., van Bergen, D.D., Rathod, S., van Straten, A.,Bhui, K., Kerkhof, A.J.F.M. (2016). Canına Kıymak ‘crushing life energy’: A qualitative study on lay and professional understandings of suicide and help-seeking among Turkish migrants in the UK and in the Netherlands. International Journal of Culture and Mental Health, 9,1-16. Eysenbach, G. (2005). The law of attrition. Journal of Medical Internet Research, 7, 1-13. Fassaert, T., Hesselink, A.E., Verhoeff, P. A. (2009a). Acculturation and use of health care services by Turkish and Moroccan migrants: a cross-sectional population-based study. BMC Public Health,9,1-9. Fassaert, T., Peen, J., van Straten, A., de Wit, M., Schrier, A., Heijnen, H., Cuijpers, P., Verhoeff, A., Beekman, A., Dekker, J. (2010). Ethnic differences and similarities in outpatient treatment for depression in the Netherlands. Psyciatric Services,61,690-697. Fassaert, T. J. L., de Witt, M.A.S., Tuinebreijer, W. C., Verhoeff, A. P., Beekman, A.T.F., Dekker, J. (2009b). Perceived need for mental health care among non-western labour migrants. Soc Psychiatry Psychiatr Epidemiol, 44,208–216. Fergusson, D., M., Lyskey, M., T. (1995). Suicide attempts and suicidal ideation in a birth cohort of 16-year- old New Zealanders. J. AM. Acad. Child Adolesc. Psychiatry,34, 1308-1317. Fergusson, D., M., Horwood, L., J., Ridder, E., M., Beautrais, A., L. (2005). Suicidal behaviour in adolescence and subsequent mental health outcomes in young adulthood. Psychological Medicine, 35, 983-993. Fogel, J., Fogel, D. E. (2005). Stigma beliefs of Asian Americans with depression in an internet sample. Can J Psychiatry 50, 470-478. Fox, A. B., Taverna, A. C., Earnshaw, V. A., Vogt, D. (2018). Conceptualizing and Measuring Mental Illness Stigma: The Mental Illness Stigma Framework and Critical Review of Measures. Stigma and Health 3, 348–376. References ― 297

Freedenthal, S. (2007). Racial disparities in mental health service use by adolescents who thought about or attempted suicide. Suicide and Life-Threatening Behavior,37,27-34. Garssen , M. , Hoogenboezem, H., Kerkhof, A.J.F.M. (2006). Zelfdoding onder migranten groepen en autochtonen in Nederland [Suicide among migrant groups and natives in the Netherlands]. Nederlands Tijd schrift voor Geneeskunde, 150, 2143 – 2149. Gary F (2005). Stigma: barrier to mental health care among ethnic minorities. Issues in Mental Health Nursing 26, 979–999. Gençöz, T., Or, P. (2006). Associated Factors of Suicide Among University Students: Importance of Family Environment. Contemporary Family Therapy,28,261-268. Georg Hsu, L.K., Wan, YM., Chang, H., Summergrad, P., Tsang Bill Y.P., Chen, H. (2008) Stigma of depression is more severe in Chinese Americans than Caucasian Americans, Psychiatry 71:3, 210-218. Givens J. L., Houston, T.K., van Voorhees, B.W., Ford, D.E., Cooper, L.A. (2007). Ethnicity and preferences for depression treatment. General Hospital Psychiatry 29, 182-191. Goldston, D. B., Molock, S. D., Whitbeck, L. B., Murakami, J. L., Zayas, L. H., Hall, G. C. N. (2008). Cultural considerations in adolescent suicide prevention and psychosocial treatment. American Psychologist, 63,14-31.http://dx.doi.org/10.1037/0003-066X.63.1.14 Gören S., Gürkan F., Tırascı Y., Özen S. (2004). Suicide in children and adolescents at a province in Turkey. Am J Forensic Med Pathol,24,214–217. Gronholm, P. C., Henderson, C., Deb, T., Thornicroft, G. (2019). Anti-stigma interventions: Theory and evidence. In: Bhugra, D., Bhui, K., Yeung Shan Wong, S., Gillman, S. E. Oxford Textbook of Public Mental Health. p.p. 411-421. Güloğlu C., Kara I. H. (2005). Acute poisoning cases admitted to a university hospital emergency department in Diyarbakir, Turkey. Hum. Exp. Toxicol, 24,49-54. Güngör, D. (2008). The meaning of parental control in migrant, sending and host communities: Adaptation or persistence? Applied Psychology. An International Review, 57, 397–416. Güngör, D., Bornstein, M. H. (2009). Gender, development, values, adaptation and discrimination in acculturating adolescents: The case of Turk heritage youth born and living in Belgium. Sex Roles, 60, 537–548. Haigh, M., Kapur, N., Cooper, J. (2015). Suicidal Behaviour Among Ethnic Minorities in England. In van Bergen, D., Montesinos, A.H. & Schouler-Ocak M. (Eds), Suicidal behaviour of immigrants and ethnic minorities in Europe. Gottingen, Germany: Hogrefe. Halter M.J. (2004). The stigma of seeking care and depression. Archives of Psychiatric Nursing 18, R 178–184. Han, B., Compton, W.M., Gfroerer, J., McKeon, R. (2014). Mental health treatment patterns among adults with recent suicide attempts in the United States. Am. J. Public Health, 104, 2359–2368. Handojo, V. (2000). Attachment styles, acculturation attitudes/behaviors, and stress among Chinese Indonesian immigrants in the United States. Dissertation Abstracts International: Section B: The Sciences and Engineering, 61(4B), 2271. Harper Shehadeh, M., E. Heim, N. Chowdhary, A. Maercker, E. Albanese. (2016). “Cultural Adaptation of Minimally Guided Interventions for Common Mental Disorders: A Systematic Review and Meta- Analysis.” JMIR Ment Health 3 (3):e44. doi: 10.2196/mental.5776. Hassan, R. (1995). Suicide Explained: The Australian Experience, Victoria: Melbourne University Press. Haugen, P. T., MsGrills, A. M., Smid, G. E., Nijdam, M. J. (2017). Mental health stigma and barriers to mental health care for first responders: A systematic review and meta-analysis. Journal of Psychiatry Research 94, 218-229. 298 ― References

Hawton, K., Kingsbury, S., Steinhardt, K., James, A., Fagg, J. (1999). Repetition of deliberate self-harm by adolescents: the role of psychological factors. J. AM. Acad. Child Adolesc. Psychiatry, 22, 369-378. Hawton, K., Witt, K. G., Taylor Salisbury, T. L., Arensman, E., Gunnell, D., Townsend, E., . . . Hazell, P. (2015). Interventions for self-harm in children and adolescents. Cochrane Database of Systematic Reviews. doi:10.1002/14651858.CD012013. Hawton, K., Witt, K. G., Taylor Salisbury, T. L., Arensman, E., Gunnell, D., Hazell, P., . . . van Heeringen, K. (2017). Psychosocial interventions for self-harm in adults. Cochrane Database of Systematic Reviews. doi: 10.1002/14651858.CD012189 Hayes, A.F. (2013). Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach. Guilford Publications. Hedges LV, Olkin I. (1985). Statistical methods for meta-analysis. Academic Press, San Diego, CA. Hegerl, U. (2016). Prevention of suicidal behavior. Dialogues in clinical neuroscience,18, 183. Heinz, A., Kluge, U. (2012). Mental health in different groups of migrants and ethnic minority within Europe and beyond: Regional and cross-national challenges and approaches in research, practice and training. European Psychiatry 27, S1-S3, https://doi.org/10.1016/S0924-9338(12)75700-0. Hekimoğlu, Y., Melez, I. S., Canturk, N., Erkol, Z. Z., Dizdar, M. G., Canturk, G, Melez, D. O., Kir, Z. A (2016). Descriptive study of female suicide deaths from 2005 to 2011 in Van city, Turkey. BMC Women`s Health,16:20. Henkel, K.E., Brown, K., Kalichman, S.C., (2008). AIDS-related stigma in individuals with other stigmatized identities in the USA: a review of layered stigmas. Soc. Personal. Psychol. Compass 2, 1586–1599. Heredia-Montesinos, A. Aicherber, M. C., Temur-Erman, S., Bromand, Z., Heinz, A., Schouler-Ocak, M. (2018). Explanatory models of suicidality among women of Turkish descent in Germany: A focus group study. Transcult Psychiatry, 22:1363461518792432. https://doi.org/10.1177/1363461518792432. Heredia-Montesinos, A. (2015). Precipitating and risk factors for suicidal behaviour among immigrants and ethnic minorities in Europe: A review of the literature. . In: van Bergen D., Heredia Montesinos, A., Scholer-Ocak, M. Suicidal Behavior of Immigrants and Ethnic Minorities in Europe. Hogrefe Publishers editors. Hogfree Publishing. Germany. pp. 27-43. Hetrick, S. E., Robinson, J., Spittal, M. J., Carter, G. (2016). Effective psychological and psychosocial approaches to reduce repetition of self-harm: A systematic review, meta-analysis and meta-regression. BMJ Open, 6. Retrieved from http://bmjopen.bmj.com/content/6/9/e011024.abstract Hickie, I.B., Davenport, D.A., Luscombe, G.M., Rong, Y., Hickie, M.L., Bell, M.I. (2007). The assessment of depression awareness and help-seeking behaviour: Experiences with the International Depression Literacy Survey. BMC Psychiatry 7:48, https://doi.org/10.1186/1471-244X-7-48. Higgins J.P.T., Altman D.G., Gøtzsche P.C., Jüni P., Moher D., Oxman A.D., Savović J., Schulz K.F., Weeks L., Sterne J.A.C. (2011). The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 343, d5928. Higgins J.P.T., Thompson S.G., Deeks J.J., Altman D.G. (2003). Measuring inconsistency in meta-analyses. BMJ 327, 557–560. Hilderink, I., van`t Land, H., Smits, C. (2009). Trendrapportage 2009. Drop-out onder allochtone GGZ-clienten. Themarapportage. Utrecht: Trimbos-Instituut. Hisli N. (1987). Beck Depresyon Envanterinin geçerliligi üzerine bir çalışma (A study on the validity of the Beck Depression Inventory)[Article in Turkish].Turkish J Psychol,6,118-122. Hjelmeland, H. (2011). Cultural context is crucial in suicide research and prevention. Crisis, 32, 61–64. References ― 299

Hjelmeland, H. (2013). Suicide Research and Prevention: The Importance of Culture in “Biological Times” In Colucci, E., Lester, D., Hjelmeland, H., Ben Park, B. C. Suicide and Culture Understanding the Context. Hogrefe Publishing, the USA. Hjelmeland, H., Knizek, B. L. (2017). Suicide and mental disorders: A discourse of politics, power, and vested interests. Death Studies, 8, 481–492. Hofstra, J., van Oudenhoven, J.P., Buunk, B.P. (2005). Attachment styles and majority members` attitudes towards adaptation strategies of immigrants. International Journal of Intercultural Relations,29,601-619. Hogg, M. A., Smith, J. R. (2007). Attitudes in social context: A social identity perspective. Eur Rev Soc Psychol, 18, 89-131. Hom, M. A., Stanley, I. H., Joiner, T. E. (2015). Evaluating factors and interventions that influence help- seeking and mental health service utilization among suicidal individuals: A review of the literature. Clinical Psychology Review, 40, 28–39. https://doi.org/10.1016/j.cpr.2015.05.006 Hunt, I.M., Robinson, J., Bickley, H., Meehan, J., Parsons, R., McCann, K.,Appleby, L. (2003). Suicides in ethnic minorities within 12 months of contact with mental health services. National clinical survey. British Journal of Psychiatry,183,155-160. Hwang W. (2006). The psychotherapy adaptation and modification framework: Application to Asian Americans. Am Psychol 61, 702–715. Hwang, W. (2009). The formative method for adapting psychotherapy (FMAP): A community-based developmental approach to culturally adapting therapy. Prof Psychol Res Pr.,40, 369-377. İbiloğlu, A. O., Atli, A., Demir, S., Gunes, M. Kaya, M. C., Bulut, M., Sir, A. (2016). The investigation of factors related to suicide attempts in Southeastern Turkey. Neuropsychiatr Dis Treat,12, 407–416. Institute of Medicine. (2002). Reducing suicide: A national imperative. Washington, DC: The National Academies Press. Ioannidis J.P.A., Patsopoulos N.A., Evangelou E. (2007). Uncertainty in heterogeneity estimates in meta- analyses. BMJ 335, 914–916. Jansen-Kallenberg, H., Schulz, H., Kluge, U.,Strehle, J., Wittchen, H-U., Wolfradt, U. et al. (2017). Acculturation and other risk factors of depressive disorders in individuals with Turkish migration backgrounds. BMC Psychiatry,17,1-12. Jimenez, D.E.., Bartels, S.J.., Gardenas, V., Alegria, M. (2013). Stigmatizing attitudes toward mental illness among racial/ ethnic older adults in primary care. Int J Geriatr Psychiatry 28: 1061–1068. Joiner T. (2005). Why People Die by Suicide?. Cambridge, MA: Harvard University Press. R Joshi, R., Alim, M., Kengne, A. P., Jan, S., Maulik, P. K., Peiris, D., Patel, A. A. (2014). Task shifting for non-communicable disease management in low and middle income countries; a systematic review. PLoS ONE, 9, e103754. Juang, L.P., Cookston, J.T. (2009). Acculturation, discrimination and depressive symptoms among Chinese American Adolescents: A longitudinal study. J Primary Prevent, 30, 475-496. KAMER Women Centre. İstersek Biter [We Can Stop This]. Berdan Matbaacılık, İstanbul, Turkey. 2011. ISBN: 978-9944-0840-1-7. Karam, E. G., Hajjar, R. V., Salamoun, M. M. Suicidality in the Arab world Part I: Community studies. The Arab Journal of Psychiatry. 2007; 18, 99–107. Kamal, Z., Loewenthal, K. M. (2002). Suicide beliefs and behavior among young Muslims and Hindus in the UK. Mental Health, Religion & Culture, 5, 111–118. https://doi.org/10.1080/13674670210141052 Karbeyaz, K., Akkaya, H., Balci, Y. (2013). Analysis of suicide deaths in a 15-year period in Eskisehir, western Anatolia, Turkey and the determination of risk factors. Ann Saudi Med, 33, 377-382. 300 ― References

Karbeyaz, K., Toygar, M., Celiker, A. (2016). Suicide among university students in Eskisehir, Turkey. Journal of Forensic and Legal Medicine, 44, 111-115. Keeble, C., Richard Law, G., Barber, S., Baxter, P.D. (2015). Choosing a method to reduce selection bias: A tool for researchers. Open J.Epidemiol, 5,155–162. Kerkhof A.J.F.M., van Spijker B.A.J. (2011) Worrying and rumination as proximal risk factors for suicidal behaviour. In: O’Connor RC, Platt S, Gordon J, editors. International Handboook of Suicide Prevention: Research, Policy and Practice.West Sussex: Wiley - Blackwell. 199–209. Kessler R.C., Bromet E.J. (2013). The epidemiology of depression across cultures. Annu Rev Public Health 34: 119–138. Khan, M. M., Ahmad, A., Khan, S. R. (2009). Female suicide rates in Ghizer, Pakistan. Suicide and Life- Threatening Behavior, 39, 227-230. Kilberg, R. (2014). Turkey`s evolving migration identity. Retrieved on 1st August 2014, from http://www. migrationpolicy.org/ Klafus, L.H., Fassaert, T.J.L., de Wit, M.A.S. (2014). Equity of access to mental health care for anxiety and depression among different ethnic groups in four large cities in the Netherlands. Soc Psychiatry Psychiatr Epidemiol,49,1139–1149. Kleiman, E.M., Anestis, M.D. (2015). Introduction to the special issue: Recent advances in suicide research: Mediators and moderators of risk and resilience. International Journal of Cognitive Therapy, 8, 95–98. Klok, J., van Tilburg, T. G., Suanet, B., Fokkema, T., Huisman, M. (2017). National and transnational belonging among Turkish and Moroccan migrants in the Netherlands: protective against loneliness. Eur J Ageing. https://doi.org/10.1007/s10433-017-0420-9 Knaevelsrud, C., J. Brand, A. Lange, J. Ruwaard, B. Wagner. (2015). “Web-based psychotherapy for posttraumatic stress disorder in war-traumatized Arab patients: randomized controlled trial.” J Med Internet Res. 17 (3):e71. https://doi.org/10.2196/jmir.3582 Kohn, L.P., Oden, T., Munoz, R.F., Robinson, A., Leavitt, D. (2002). Brief report: Adapted cognitive behavioral group therapy for depressed low-income African American women. Community Mental Health Journal, 38,497-504. Konkan, R., Erkus, G. H., Guclu, O., Senormanci, O., Aydin, E., Ulgen, M. C., Sungur, M. Z. (2014). Coping Strategies in Patients Who Had Suicide Attempts. Archives of Neuropsychiatry, 51, 46-51. Korszun, A., Sokratis, D., Ahmed, K. and Bhui, K. (2012). Medical student attitudes about mental illness: Does medical-school education reduce stigma? Academic Psychiatry 36:197-204. Köse, A., Eraybar, S., Kose, B., Koksal, O., Aydin, S. A., Armagan, E., Ozdemir, F. (2012). Patients Over the Age of 15 Years Admitted for Attempted Suicide to the Emergency Department and the Psychosocial Support Unit. The Journal of Academic Emergency Medicine, 11, 193-196. Koslofsky S., Rodriguez D.M.M. (2017). Cultural adaptations to psychotherapy: Real-world applications. Clin Case Stud 16: 3-8. Kuester, L., Eylem, O., Melissourgaki, Z., Freestone, M., Dein, S., Bhui, K. (2019). Interventions for Mental Health Across Cultures: Post-conflict migration and gender-related mental health challenges. Inform Final Project Report. Centre for Protection of National Infrastructure. London, UK. Kushner M., Sher K. (1991). The relation of treatment fearfulness and psychological service utilization: an overview. Professional Psychology: Research and Practice,22, 196–203 Lai, M. H., Maniam, T., Chan, L. F., Ravindran, A. V. (2014). Caught in the web: A review of web- based suicide prevention. Journal of Medical Internet Research, 16(1),e30. https://doi.org/10.2196/ jmir.2973. References ― 301

Lamers L.M., Stalmeier P.F., McDonnell J., Krabbe P.F., van Busschbach J.J. (2005) Measuring the quality of life in economic evaluations: the Dutch EQ-5D tariff [Article in Dutch]. Ned Tijdschr Geneeskd.9, 149,1574-1578. Lavrakas, P. J. (2008). Encyclopedia of survey research methods Thousand Oaks, CA: Sage Publications, Inc. http://dx.doi.org/10.4135/9781412963947.n523. Lawrie, S., M., McIntosh, A., M., Rao, S. (2000). Critical Appraisal for Psychiatry (pp.5-10) Elsevier, Churchill Livingstone. Le Cook, B., Trinh, N-H., Li, Z., Hou, S-H., S., Progovac, A. (2017). Trends in racial-ethnic disparities in access to mental health care, 2004–2012. Psychiatr Serv. 68, 9–16. Leigh, S., Flatt, S. (2015). App-based psychological interventions: Friend or foe? Evidence Based Mental Health, 18, 97. Retrieved from http://ebmh.bmj.com/content/18/4/97.abstract Lee S., Juon HS, Martinez G., Hsu, C.E., Robinson, E.S., Bawa, J., Ma, G. (2009). Model minority at risk: expressed needs of mental health by Asian American young adults. J Community Health 34:144–152. Lester, D. (2014). Opression and Sucicide. Suicidology Online. 2014; 5: 59-73. Lester, D. (2006). Suicide and Islam. Achieves of Suicide Research,10,77-97. Lester, D. (2008). Suicide and culture. World Cultural Psychiatry Research Review, 3,51-68. Lester, D. (2013). The cultural meaning of suicide. What does it mean? In D. Lester, E. Colucci (Eds.), Suicide and culture: Understanding the context. Gottingen, Germany: Hogrefe Publishers. Lester, D. (2015). Suicide Among the Roma and Irish Travellers In van Bergen, D., Montesinos, A.H., Schouler-Ocak, M. (2015). Suicidal behavior of immigrants and ethnic minorities in Europe. Hogrefe Publishers. Lester, D., (2013). Culture and Suicide. In Colucci, E., Lester, D., Hjelmeland, H. & Park, B. (2013). Suicide and Culture: Understanding the Context. Hogrefee, USA. Lester, D., Beck, A. T., Mitchell, B. (1979). Extrapolation from attempted suicides to completed suicides: A test. Journal of Abnormal Psychology, 88,78-80. Lewin S.A., Dick J., Pond P., Zwarenstein M., Aja G., van Wyk B., Bosch- Capblanch X., Patrick M. (2005) Lay health workers in primary and community health care: Cochrane systematic review. International Journal of Epidemiology, 34,1250-1251. Lindert, J., Ehrenstein, O. S., Priebe, S., Mielck, A., Brahler, A. (2009). Depression and anxiety in labor migrants and refugees-a systematic review and meta-analysis. Social Science and Medicine,69,246-257. Lindert, J., Schouler-Ocak, M., Heinz, A., Priebe, S. (2008). Mental health, health care utilisation of R migrants in Europe. European Psychiatry,23,14-20. Linehan M.M. (1993) Cognitive-Behavioral Treatment of Borderline Personality Disorder. New York: Guilford Press. 29. Linehan M.M. (1993) Skills Training Manual for Treating Borderline Personality Disorder. New York: Guilford Press. Lipson, J.G., Meleis, A.I. (1989). Methodological issues in research with immigrants. Medical Antropology,12,103-115. Lizardi, D., Grunebaum, M. F., Burke, A., Stanley, B., Mann, J. J., Harkavy – Friedman, J. et al. (2006). Epidemiologie suizidalen verhaltens von migranten in Deutschland [Epidemiology of suicidal behavior of migrants in Germany]. Suizidprophylaxe,33,171- 176. Lizardi, D., Grunebaum, M.F., Burke, A., Stanley, B., Mann, J.J., Harkavy-Friedman, J. and Oquendo, M. (2011). The effect of social adjustment and attachment style on suicidal behaviour. Acta Psychiatr Scand,124, 295–300. 302 ― References

Logie, C.H., James, L.L., Tharao, W., Loutfy, M.R. (2011). HIV, gender, race, sexual orientation, and sex work: a qualitative study of intersectional stigma experienced by HIV-positive women in Ontario, Canada. PLoS Med. 8, e1001124. Löhr, C., Schmidtke, A., Wohner, J., Sell, R. (2006). Epidemiologie suizidalen Verhaltens von Migranten in Deutschland. . Suizidprophylaxe 4, 171-176. Lucksted, A., Drapalski, A., Calmes, C., Forbes, C., DeForge, B., Boyd, J. (2011). Ending self-stigma: Pilot evaluation of a new intervention to reduce internalized stigma among people with mental illnesses. Psychiatric Rehabilitation Journal, 35, 51–54. http://dx.doi.org/10.2975/35.1 .2011.51.54 Makowski, AC., von dem Knesebeck, O. (2017). Depression stigma and migration – results of a survey from Germany. BMC Psychiatry 17:381, http://dx.doi.org/10.1186/s12888-017-1549y. Marecek, J. (1998). Culture, Gender, and Suicidal Behavior in Sri Lanka. Suicide and Life-Threatening Behavior, 28, 69-81. Martinez, R., Whitfield, G., Dafters, R., Williams, C. (2007). Can people read self-help manuals for depression? A challenge for the stepped care model and book prescription schemes. Behavioural and Cognitive Psychotherapy, 36, 89-97. Mcguire, T.G., Miranda, J. (2008). Racial and ethnic disparities in mental health care: evidence and policy implications. Health Aff (Millwood). 27, 393–403. McKenzie, K., Bhui, K. (2007). Instititional racisim in mental health care. BMJ, 334,649-650. Menke, R., Flynn, H. (2009). Relationships between stigma, depression, and treatment in White and African American primary care patients. The Journal of Nervous and Mental Disease 197, 407-411. Miranda, J., McGuire, T.G., Williams, D.R., Wang, P. (2008). Mental health in the context of health disparities. Am J Psychiatry, 9, 1102-1108. Miranda, R., Soffer, A., Polanco-Roman, L., Wheeler, A., Moore, A. (2015). Mental health treatment barriers among racial/ethnic minority versus White young adults 6 months after intake at a college counseling center. Journal of American College Health 63, 291-299. Mishara, B., Kerkhof, A. (2013). Suicide prevention and new technologies: Evidence based practice. Basingstoke, UK: Palgrave Macmillan. Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., PRISMA Group (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine. Mokkarala, S., O`Brien, E.K., Siegel, J.T. (2016). The relationship between shame and perceived biological origins of mental illness among South Asian and white American young adults. Psychology, Health & Medicine, http://dx.doi.org/10.1080/13548506.2015.1090615. Mooren T., Knipscheer J., Kamperman A., Kleber R., Komproe I. H. (2001). The Lowlands Acculturation Scale. Validity of an Adaptation Measure Among Migrants in the Netherlands. In The impact of war. Studies on the psychological consequences of war and migration. Edited by: Mooren T. Delft: Eburon Publishers,49-70. Morgan, C., Mallett, R., Hutchinson, G., Leff, J. (2009). Negative pathways to psychiatric care and ethnicity: The bridge between social science and psychiatry. Social Science and Medicine, 58, 739-752; https://doi:10.1016/S0277-9536(03)00233-8. Mugisha, J., Hjelmeland, H., Kinyada, E., Knizek, B.L. (2012). The internal dialogue between the individual and community. A discourse analysis of public views on suicide among the Baganda, Uganda. International Journal of Culture and Mental Health,5,1-15. Murray, R.L., Bauld, L., Hackshaw, L.E., McNeill, A. (2009). Improving Access to smoking cessation services for disadvantaged groups: a systematic review. Journal of Public Health,31, 258-277. References ― 303

Mühlmann, C., Madsen, T., Hjorthoj, C., Kerkhof, A.J.F.M., Nordentoft, M. and Erlangsen, A. (2017). The Self-help Online against Suicidal thoughts (SOS) trial: study protocol for a randomized controlled trial. Tr ials, 18:45. https://doi.org/10.1186/s13063-017-1794-x Nadeem, E., Lange, J.M., Edge, D., Fongwa, M., Belin, T., Miranda, J. (2007). Does stigma keep poor young immigrant and US-Born Black and Latina women from seeking mental health care? Psychiatric Services,58,1547-1556. National Institute for Health and Care Excellence (2014). Psychosis and schizophrenia: treatment and management.(Clinical guideline 178). Retrieved February 25,2016, from http://guidance.nice.org.uk/ CG178. National Collaborating Centre for Mental Health (2019). Self-Harm: The Short Term Physical and Psychological Management and Secondary Prevention of Self-Harm in Primary and Secondary Care. Clinical Guideline 16. London: Gaskell & British Psychological Society. Nock MK, Borges G, Bromet EJ, et al. (2008). Suicide and suicidal behavior. Epidemiol Rev,30,133-54. https://doi.org/10.1093/epirev/mxn002 Noh S., Kaspar V. (2003) Perceived discrimination and depression: moderating effects of coping, acculturation, and ethnic support. Am J Public Health,93, 232–238. O`Brien, R., Fahmi, R., Singh, S.P. (2009), Disengagement from mental health services: A literature review. Soc Psychiatry Psychiatr Epidemiol, 44,558–568. O’Carroll, P. W., Berman, A. L., Maris, R. W., Moscicki, E. K., Tanney, B. L., Silverman, M. M. (1996). Beyond the Tower of Babel: A nomenclature for suicidology. Suicide and Life Threatening Behavior, 26, 237-252. O`Conner, K., Copeland, VC., Grote, NC., Koeske, G., Reynolds, CF., Brown, C. (2010). Mental health treatment seeking among older adults with depression: The impact of stigma and race. Am J Geriatr Psychiatry 18:6,531-543. O`Conner, K., Koeske, G., Brown, C. (2009) Racial Differences in Attitudes Toward Professional Mental Health Treatment: The Mediating Effect of Stigma, Journal of Gerontological Social Work, 52:7, 695-712, https://doi.org/10.1080/01634370902914372. O’ Connor, R. C. (2016). The international handbook of suicide prevention. John Wiley & Sons. Okasha, A., Okasha, T. (2009). Suicide and Islam. In D. Wasserman, C. Wasserman (Eds.), Oxford textbook of suicidology and suicide prevention: A global perspective (pp.49–56). Oxford, UK: Oxford University Press. O`Mahen, H.A., Henshaw, E., Jones, J.M., Flynn, H.A. (1999). Stigma and depression during pregnancy, R does race matter? The Journal of Nervous and Mental Disease 199, 257-262. Öner S., Yenilmez C., Ayranci U., Gunay Y., Ozdamar K. (2007). Sexual differences in the suicides in Turkey.Eur Psychiatry,22,223-228. Öner, S., Yenilmez, C., Ozdamar, K. (2015). Sex-related differences in methods of and reasons for suicide in Turkey between 1990 and 2010. Journal of International Medical Research, 43, 483–493. Oyserman, D., Lee, S. W. (2008). Does culture influence what and how we think? Effects of priming individualism and collectivism. Psychological Bulletin, 134, 311–342. Özdel, O., Varma, G., Atesci, F. C., Oguzhanoglu, N. K., Karadag, F., Amuk, T. (2009). Characteristics of suicidal Behavior in a Turkish sample. Crisis,30,90–93. Özdemir, S., Mutluer, S.N., Özyürek, E. (2019) Exile and plurality in neoliberal times: Turkey’s Academics for Peace. Public Culture, 31, ISSN 0899-2363 (In Press). Papadopoulos, C., Leavey, G., Vincent, C. (2002). Factors influencing stigma A comparison of Greek- Cypriot and English attitudes towards mental illness in north London. Soc Psychiatry Psychiatr Epidemiol 37,430–434. 304 ― References

Pascoe EA, Smart Richman L. (2009) Perceived discrimination and health: a meta-analytic review. Psychol Bull, 135,531–554. Patel, S.P., Gaw, A.C. (1996). Suicide among immigrants from the Indian subcontinent: A review. Psychiatric Services,47,517-521. Patel, V., Weiss, H. A., Chowdhary, N., Naik, S., Pednekar, S., Chatterjee, S.,…Kirkwood, B. R. (2010). Effectiveness of an intervention led by lay health counsellors for depressive and anxiety disorders in primary care in Goa, India (MANAS): a cluster randomised controlled trial. Lancet, 376, 2086–2095. Patel, V., Weobong, B., Weiss, H. A., Anand, A., Bhat, B., Katti, B.,…Fairburn, C. G. (2017). The Healthy Activity Program (HAP), a lay counsellor-delivered brief psychological treatment for severe depression, in primary care in India: A randomised controlled trial. Lancet, 389, 176–185. Perry, A., Gardener, C., Dove, C., Eiger, Y., Loewenthal. (2018). Improving mental health knowledge of the Charedi Orthodox Jewish Community in North London: A partnership project. International Journal of Social Psychiatry, Vol. 64(3) 235–247 Perry, A., Gardener, C., Oliver, J.E., Tas, C., Ozenc, C. (2019). Exploring the cultural flexibility of the ACT model as an effective therapeutic group intervention for Turkish speaking communities in East London. The Cognitive Behaviour Therapist,12, 1 -25 https://doi.org/10.1017/S1754470X1800004 Pescosolido, B. A., Martin, J. K., Lang, A., Olafsdottir, S. (2008). Rethinking theoretical approaches to stigma: A framework integrating normative influences on stigma (FINIS). Social Science & Medicine, 67, 431–440. Pettigrew, T. F., Tropp, L. (2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90, 751–783. Phalet, K., Schonpflug, U. (2011). Intergenerational transmission of collectivism and achievement values in two acculturation contexts. The case of Turkish families in Germany and Turkish and Moroccan families in the Netherlands. Journal of Cross Cultural Psychology,32,186-201. Phalet, K., Baysu, G., van Acker, K. (2015). Ethnicity and migration in Europe. International Encyplopedia of Social and Behavioural Sciences (Second Edition),142-147. Phalet, K., Schönpflug, U. (2001). Intergenerational transmission of collectivism and achievement values in two acculturation contexts: The case of Turkish families in Germany and Turkish and Moroccan families in the Netherlands. Journal of Cross Cultural Psychology, 32, 186–201. Picco, L., Panga, S., Laua, Y.W., Jeyagurunathana, A., Satgharea, P., Abdina, E., Vaingankara, J.A., Limb, S., Pohc, C.L., Chonga, S.A., Subramaniam, M. (2016). Internalized stigma among psychiatric outpatients: Associations with quality of life, functioning, hope and self-esteem. Psychiatry Research 246, 500-506. Polanco-Roman, L., Miranda, R. (2013). Culturally-related stress, hopelessness, and vulnerability to depressive symptoms and suicidal ideation in emerging adulthood. Behavior Therapy,44,75-87. Polatöz, Ö., Kuğu, N., Doğan, O., Akyüz, G. (2010).Sivas il merkezinde intihar davranışının yaygınlığı ve bazı sosyo demografik faktörlerle ilişkisi [Article in Turkish]. Düşünen Adam Psikiyatri ve Nöroloji Bilimler Dergisi,24,12-23. Poppleton, A., Gire, N. (2017). mHealth: Bridging the mental health gap in central and eastern Europe. The Lancet Psychiatry, 4, 743-744. doi:10.1016/S2215-0366(17)30359-0 Rabin R., Charro F.D. (2001). EQ-5D: a measure of health status from the EuroQuol group. Ann Med,33,337-343. Radcliffe, J., Doty, N., Hawkins, L.A., Gaskins, C.S., Beidas, R., Rudy, B.J. (2010). Stigma and sexual health risk in HIV-positive African American young men who have sex with men. AIDS Patient Care STDS 24, 493–499. References ― 305

Rahman, A., Hamdani, S. U., Awan, N. R., Bryant, R. A., Dawson, K. S., Khan, M. F.,…van Ommeren, M. (2016). Effect of a multicomponent behavioral intervention in adults impaired by psychological distress in a conflict-affected area of Pakistan: A randomized clinical trial. JAMA, 316, 2609–2617. Rahman, A., Malik, A., Sikander, S., Roberts, C., Creed, F. (2008a). Cognitive behaviour therapy-based intervention by community health workers for mothers with depression and their infants in rural Pakistan: A cluster-randomised controlled trial. Lancet, 372, 902–909. Rao, D., Feinglass, J., Corrigan, P. (2007). Racial and ethnic disparities in mental illness stigma. The Journal of Nervous and Mental Disease 195, 1020-1023. Rathod S., Kingdon D., Pinninti N., Turkington D., Phiri P. (2015). Cultural Adaptation of CBT for Serious Mental illness: A guide for training and practice. Wiley – Blackwell. Rathod, S., Kingdon, D., Phiri, P., Gobbi, M. (2010). Developing culturally sensitive cognitive behaviour therapy for psychosis for ethnic minority patients by exploration and incorporation of service users` and health professionals` views and opinions. Behavioural and Cognitive Psychotherapy,38,511-533. Rathod, S., Persaud, A., Naeem, F., Pinninti, N., Tribe, R., Eylem, O., Pihiri, P., Muzaffar, S., Irfan, M. (2019a). Culturally adapted interventions in mental health: Global position statement. World Cultural Psychiatry Research Review. Rathod, S., Phiri, P., Harris, S., Underwood, C., Thagadur, M., Padmanabi, U., Kingdon, D. (2013). Cognitive behaviour therapy for psychosis can be adapted for minority ethnic groups: a randomized controlled trial. J Schizophr Res, 143, 319–326. Rathod, S., Phiri, P., Naeem, F. (2019b). An evidence-based framework to culturally adapt cognitive behaviour therapy. The Cognitive Behaviour Therapist, 12, e10, 1-15. http://dx.doi.org/10.1017/ S1754470X18000247 Razum, O., Zeeb, H. (2004). Suicide mortality among Turks in Germany. [Article in German] Nervenarzt, 75, 1092-1098 Razum, O., Zeeb, H., Akgun, H.S., Yilmaz, S. (1998). Low overall mortality of Turkish residents in Germany persists and extends into second generation: Merely a healthy migrant effect? Tropical Medicine International Health, 3, 297–303. Rezaeian, M. (2010). Suicide Among Young Middle Eastern Muslim Females The Perspective of an Iranian Epidemiologist. Crisis, 31, 36–42. Richards D., Richardson T. (2012). Computer-based psychological treatments for depression: A systematic review and meta-analysis. Clin Psychol Rev. 32: 329–342. Rimkeviciene, J., O’Gorman, J., Hawgood, J., De Leo, D. (2019). Development and validity of the Personal R Suicide Stigma Questionnaire (PSSQ): A New Tool to Assess Stigmatization Among Those Who Are Suicidal. Crisis,The Journal of Crisis Intervention and Suicide Prevention. Advance online publication. http://dx.doi.org/10.1027/0227-5910/a000567 Roberts, T., Esponda, G.M., Krupchanka, D., Patel, V., Rathod, S. (2018). Factors associated with health service utilisation for common mental disorders: a systematic review. BMC Psychiatry, 18:262. https://doi.org/10.1186/s12888-018-1837-1 Roccas, S., Horenczyk, G., Schwartz, S. H. (2000). Acculturation discrepancies and well- being: the moderating role of conformity. European Journal of Social Psychology, 30, 323-334. Rossouw, T.I. (2013). Mentalization based treatment: Can it be translated into practice in clinical settings and teams? Journal of the American Academy of Child and Adolescent Psychiatry,52,220-222. Rossouw, T. I., Fonagy, P. (2012). Mentalization-based treatment for self-harm in adolescents: A randomized controlled trial. J Am Acad Child Adolesc Psychiatry,51, 1304-1313. Rüsch, N., Evans-Lacko, S., Thornicroft, G. (2012). What is a mental illness? Public views and their effects on attitudes and disclosure. Australian & New Zealand Journal of Psychiatry 46, 641 –650. 306 ― References

Salant, T., Lauderdale, D.S. (2003). Measuring culture: a critical review of acculturation and health in Asian immigrant populations. Social Science & Medicine, 57, 71-90. Saraçoğlu, U., Gokel, Y., Ay, M. O., Avci, A., Eroglu, M. Z., Kara, M. E., Canacankatan, M., Dogan, M., Acehan, S. (2014). İlaç alımı yoluyla ozkıyım girişimleri [Suicide attempts via drug intake] Medical Journal of Bakırköy,40, 18-23. Sayil, I., Devrimci-Ozguven, H. (2002). Suicide and suicide attempts in Ankara in 1998: Results of the WHO/EURO multicenter study of suicidal behaviour,Crisis,1,11-16. Schafer, T., Wood, S., Williams, R. (2011). A survey into student nurses` attitudes towards mental illness: Implications for nurse training. Nurse Education Today 31, 328-332. Schmitt, M.T., Branscombe, N.R. (2002). The meaning and consequences of perceived discrimination in disadvantaged and privileged social groups. European Review of Social Psychology,12,167-199. Scholer-Ocak, M. (2015) End your silence, not your life: A suicide prevention campaign for women of Turkish origin in Berlin. In: van Bergen D., Heredia Montesinos, A., Scholer-Ocak, M. Suicidal Behavior of Immigrants and Ethnic Minorities in Europe. Hogrefe Publishers editors. Hogfree Publishing. Germany. p. 173-185. Schomerus G., Angermeyer M. (2008). Stigma and its impact on help-seeking for mental disorders: what do we know? Epidemiologia e Psichiatria Sociale 17,31–37. Schünemann, H., Brożek, J., Guyatt, G., Oxman, A. (2013). GRADE Handbook. Cochrane Collaberative Training. Schweitzer, R., Melville, F., Steel, Z., Lacherez, P. (2006). Trauma, post-migration living difficulties, and social support as predictors of psychological adjustment in resettled Sudanese refugees. Australian and New Zealand Journal of Psychiatry,40,179-187. Seber G, Dilbaz N, Kaptanoğlu C., Tekin D. (1993). Umutsuzluk ölçeği: geçerlilikvegüvenirliği (The Hopelessness Scale: Validity and reliability) [Article in Turkish]. Kriz Dergisi,1,139-142. Segal Z.V., Williams J.M.G., Teasdale J.D. (2002). Mindfulness-based Cognitive Therapy for Depression: A new approach to preventing relapse. New York: Guilford Press. Selby, E.A., Joiner, T. E., Jr. (2009). Cascades of emotion: The emergence of borderline personality disorder from emotional and behavioral dysregulation. Behaviour Research and Therapy, 46, 593–611 Selkirk, M., Quayle, E., Rothwell, N. (2014). A systematic review of factors affecting migrant attitudes towards seeking psychological help. Journal of Health Care for the Poor and Underserved, 25, 94-127. Şenol, V., Unalan, D., Avsarogullari, L., Ikizceli, I. (2005). İntihar girişimi nedeniyle Erciyes Üniversitesi Tıp Fakültesi Acil Anabilim Dalı’na başvuran olguların incelenmesi [An analysis of patients admitted to the Emergency Department of Erciyes University Medical School due to suicidal attempt]. Anadolu Psikiyatri Dergisi, 6,19-29. Shamblaw, A.L., Botha, F.B., Dozois, D.J.A. (2015). Accounting for differences in depression stigma between Canadian Asians and Europeans. Journal of Cross-Cultural Psychology 46, 597 –611. Shand, F.L., Ridani, R., Tighe, J. and Christensen, H. (2013). The effectiveness of a suicide prevention app for indigenous Australian youths: Study protocol for a randomised controlled trial. Tr ials, 14:396. Sheehan, A.E., Walsh, R.F.L., Richard, T.L. (2018). Racial and ethnic differences in mental health service utilization in suicidal adults: A nationally representative study. Journal of Psychiatric Research,107,114–119. Shefer, G., Rose, D., Nellums, L., Thornicroft, G., Henderson, C., Evans-Lacko, S. O. (2012). Our community is the worst’: The influence of cultural beliefs on stigma, relationships with family and help-seeking in three ethnic communities in London. Int J Soc Psychiatry, 59, 535–544. References ― 307

Sheftall, A.H. (2010). Attachment and Suicidality in Adolescents: An Exploration of Mediators and Moderators. Dissertation. The Ohio State University. The United States. Sher, L., Vilens, A. (2010). Immigration and mental health: Stress, psychiatric disorders and suicidal behaviour among immigrants and refugees. New York, NY: Nova Science Publishers. Shim, R.T., Compton, M.T., Rust, G.,Gruss, D.J., Kaslow, N.J. (2009). Race-ethnicity as a predictor of attitudes toward mental health treatment seeking. Psychiatric Services,60,10-16. Şimsek, Z.. Demir, C., E.R.G., Munir, M.K. (2013). Evaluation of attempted suicide in emergency departments in Sanliurfa province, Southeastern Turkey. Z Gesundh Wiss,21, 325–331. Snowden L.R., Yamada A.M. (2005). Cultural differences in access to care. Annu Rev Clin Psychol, 1,143–166. Spinhoven P., Ormel J., Sloekers P.P, Kempen G.I., van Hemert A.M. (1997). A validation study of the Hospital Anxiety and Depression Scale (HADS) indifferent groups of Dutch subjects. Psych med, 27,363-370. Stanley, B., Brown, G. K. (2012). Safety Planning Intervention: A Brief Intervention to Mitigate Suicide Risk. Cogn Behav Pract, 19, 256-264. Steel, Z., Marnane, C., Iranpour, C., Chey, T., Jackson, J.W., Patel, V., Silove, D. (2014). The global prevalence of common mental disorders: a systematic review and meta-analysis 1980-2013. Int J Epidemiol 43, 476-93. Stepp, S.D., Morse, J.Q. , Kristen, Y.E., Reynolds, S.K., Reed, I.L., Pilkonus, P.A. (2008). The role of attachment styles and interpersonal problems in suicide-related behaviors. Suicide Life Threat Behav 38,592. https://doi.org/10.1521/suli.2008.38.5.592 Stevens, W., Thorogood, M., Kayikki, S. (2002). Cost-effectiveness of a community anti-smoking campaign targeted at a high risk group in London. Health Promotion International,17, 43-50. Stöber J., Bittencourt J. (1998) Weekly assessment of worry an adaptation of the Penn State Worry Questionnaire for monitoring changes during treatment. Behav Res Ther, 36,645-656. Subramaniam, M., Abdin, E., Picco, L., Pang, S., Shafie S., Vaingankar, J.A., Kwok K.W., Verma K., Chong, S.A. (2017). Stigma towards people with mental disorders and its components – a perspective from multi-ethnic Singapore. Epidemiology and Psychiatric Sciences 26, 371–382. Sümer, N., Güngör, D. (1999). Yetişkin baglanma stilleri ölçeklerinin Türk örneklemi üzerinde psikometrik degerlendirmesi ve kültürler arası bir karşılaştırma [Psychometric assessment of the adult attachment scale on Turkish sample and a cross cultural comparison]. Türk Psikoloji Dergisi, 14, 71–106. Taktak, S., Uzun, I., Balcioglu, I. (2013). Gender differences in suicides in Istanbul, Turkey. Journal of R Affective Disorders,145,394-399. Taş, N., Guden, M., Tekin, H., Guler, I, Doner, F., Kalen, G. (2008). Voice of Men. Mental Health Needs Assessment of Turkish/Kurdish and Cypriot/Turkish Men in Hackney. Department of Mental Health. Thornicroft G. (2008). Stigma and discrimination limit access to mental health care. Epidemiologia e Psichiatria Sociale 17,14–19. Thornicroft, G., Chatterji, S., Evans-Lacko, S., Gruber, M., Sampson, N., Aguilar-Gaxiola, S., Al- Hamzawi, A., Alonso, J., Andrade, L., Borges, G., Bruffaerts, R., Bunting, B., de Almeida, J. M.C., Florescu, S., de Girolamo, G., Gureje, O., Haro, J.M., He, Y., Hinkov, H., Karam, E., Kawakami, N., Lee, S., Navarro-Mateu, F., Piazza, M., Posada-Villa, J., de Galvis, Y.T., Kessler, R. C. (2017). Undertreatment of people with major depressivedisorder in 21 countries. The British Journal of Psychiatry,210, 119–124. 308 ― References

Till, A., Bhugra, D. (2015). Suicidal behaviour and sociocultural factors among South Asians in the UK In van Bergen, D., Montesinos, A.H., Schouler-Ocak, M. (2015). Suicidal behavior of immigrants and ethnic minorities in Europe. Hogrefe Publishers. Townsend E., Hawton K., Altman D.G., Arensman E., Gunnell D., et al. (2001) The efficacy of problem- solving treatments after deliberate self-harm: meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Psychol Med 31, 979–88. Tseng, W.S., J. Streltzer, eds (2004). Cultural competence in clinical psychiatry. Washington, D.C. : American Psychiatric Publishing Inc. Tseng, W-S., Wu, D.Y.H. (1985).Chinese Culture and Mental Health. Academic Press INC. Harcourt Brace Jovanovich Publishers. Tucker, J. R., Hammer, J. H., Vogel, D. L., Bitman, R. L., Wade, N. G., Maier, E. J. (2013). Disentangling self-stigma: Are mental illness and help-seeking self-stigmas different? Journal of Counseling Psychology, 60, 520–531, http://dx.doi.org/10.1037/a0033555. Tuna, E., Bozo, O. (2014). Exploring the link between emotional and behavioral dysregulation: A test of emotional cascade model. The Journal of General Psychology,1,1-17. Turhan E., Inandi T., Aslan M., Zeren C. (2011). Epidemiology of attempted suicide in Hatay, Turkey. Neurosciences (Riyadh),16,347-52. Turner, B. J., Austin, S. B., Chapman, A. L. (2014). Treating nonsuicidal self-injury: A systematic review of psychological and pharmacological interventions. Canadian Journal of Psychiatry, 59, 576-585. Türkiye Statistik Kurumu (TÜİK) intahar statistikleri, 2017 [Turkish National Institute of Statistics on suicide statistics, 2017]. http://www.tuik.gov.tr. Accessed 17 September 2018. Usser, J. M. (2010). Are we medicalising women`s misery? A critical review of women`s higher rates of reported depression. Feminism & Psychology, 20:9. Ünlü İnce, B., Cuijpers, P., van`t Hof, E., Wouter, B., Christensen, H., Riper, H. (2013). Internet-based, culturally sensitive, problem-solving therapy for Turkish migrants with depression: Randomized controlled trial. Journal of Medical Internet Research, 15, e227. Ünlü İnce, B., Fassaert, T., de Wit, M.A.S., Cuijpers, P., Smit, J., Ruwaard, J. et al. (2014a). The relationship between acculturation strategies and depressive and anxiety disorders in Turkish migrants in the Netherlands. BMC Psychiatry,14,252. Retrieved from http://www.biomedcentral.com/1471- 244X/14/252 Ünlü İnce, B., Cuijpers, P., van`t Hof E., Riper, H. (2014b). Reaching and recruiting Turkish migrants for a clinical trial through Facebook: A process evaluation. Retrieved on the 1st August 2014 from http:// dx.doi.org/10.1016/j.invent.2014.05.003. Ünlü İnce, B., Riper, H., van`t Hof, E. Cuijpers, P. (2014c). The effects of psychotherapy on depression among racial ethnic minority groups: a metaregression analysis. Psychiatric Services in Advance, 1-6. Üstün, T.B., Ayuso-Mateos, J.L., Chatterjl, S., Mathers, C., Murray, C.J.L., Murr, C.J.L. (2004). Global burden of depressive disorders in the year 2000. British Journal of Psychiatry, 184-392. Van Acker, K., Vanbeselaere, N. (2011). Heritage culture maintenance precludes host culture adoption and vice versa. Fleming`s perceptions of Turks` acculturation behavior. Group Processes & Intergroup Relations,15,133-145. Van Beek, W. (2013). Future Thinking in Suicidal Patients: Development and Evaluation of a Future Oriented Group Training in a Randomized Controlled Trial. Print Service Ede BV, Ede, The Netherlands. Van Bergen, D. D., Smit, J. H., van Balkom, A. J., Saharso, S. (2009). Suicidal behaviour of young immigrant women in the Netherlands: Can we use Durkheim`s concept of ‘’fatalistic suicide’’ to explain their high incidence of attempted suicide? Ethnic and Racial Studies,32,301-322. References ― 309

Van Bergen, D., Eikelenboom, M., van de Looij-Jansen, P. P. (2018). Attempted suicide of ethnic minority girls with a Caribbean and Cape Verdean background: rates and risk factors. BMC Psychiatry,18:14. Van Bergen, D., Eylem, O., Tasdelen, A., Heredıa-Montesinos, A. (2019). Attempted suicide and suicide of young Turkish women in Europe and Turkey. A systematic literature review of characteristics and precipitating factors. Manuscript Submitted for Publication. Van Bergen, D.D., Montesinos, A.H., Schouler-Ocak, M. (2015). Suicidal Behavior of Immigrants and Ethnic Minorities in Europe. Hogrefe Publishers. Van Bergen, D.D., Smit, J.H., Van Balkam, A.J.L.M., Van Ameijden, E., Saharso, S. (2008). Suicidal ideation in ethnic minority and majority adolescents in Utrecht, the Netherlands.Crisis,29, 202-208. Van Bergen, D.D., van Balkom, A.J., Smit, J.H., Saharso, S. (2012). ‘’I felt so hurt and lonely’’: Suicidal behaviour in South Asian-Surinamese, Turkish and Moroccan women in the Netherlands. Transcultural Psychiatry, 49,69-86. http://dx.doi.org/10.1177//1363461511427353 Van der Stuyft, P., De Muynck, A.,Schillemans, L., Timmerman, C. (1989). Migration, acculturation and utilization of primary mental health care. Soc. Sci. Med., 29, 53-60. Van Ginneken, N., Tharyan, P., Lewin, S., Rao, G. N., Meera, S. M., Pian, J., Patel, V. (2013). Non- specialist health worker interventions for the care of mental, neurological and substance-abuse disorders in low- and middle-income countries. Cochrane Database of Systematic Reviews, 11, CD009149. Van Hemert A., Kerkhof A., de Keijser J., Verwey B. (2012). Multidisciplinaire richtlijn voor diagnostiek en behandeling van suïcidaal gedrag. Utrecht: Nederlandse Vereniging voor Psychiatrie/Nederlands Intstituut voor Psychologen/Trimbos Instituut. Van Leeuwen, N., Rodgers, R., Régner, I. and Chabrol, H. The role of acculturation is suicidal ideation among second-generation immigrants adolescents in France. Transcultural Psychiatry, 47,2010, 812-833. Van Oudenhoven, J.P., Hofstra, J. (2006). Personal reactions to ‘strange situations’: Attachment styles and acculturation attitudes of immigrants and majority members. International Journal of Intercultural Relations,30,783-798. Van Spijker B.A.J., van Straten A., Kerkhof A.J.F.M. (2010). The effectiveness of a web based self-help intervention to reduce suicidal thoughts: a randomized controlled trial. Trials, 11:25. Van Spijker B.A.J., van Straten A., Kerkhof A.J.F.M. (2014) Effectiveness of Online Self-Help for Suicidal Thoughts: Results of a Randomised Controlled Trial. PLoS ONE 9(2): e90118. http://dx.doi. org/10.1371/journal.pone.0090118 Van Spijker, B. A. J. (2012). Reducing the Burden of Suicidal Thoughts through Online Help. Ipskamp R Drukkers, Enschede. Van Spijker, B. A. J., Batterham, P. J., Calear, A. L., Farrer, L., Christensen, H., Reynolds, J., Kerkhof, A. J. M. (2014). The suicidal ideation attributes scale (SIDAS): Community-Based validation study of a new scale for the measurement of suicidal ideation. Suicide and Life-Threatening Behavior. 1-12. Van Spijker, B. A. J., van Straten A., Kerkhof, A.J. F. M. (2014). Effectiveness of online self-help for suicidal thoughts: results of a randomised controlled trial. Plos 1,9,1-8. Van Spijker, B.A. J., van Straten, A., Kerkhof, A. J. F. M., Hoeymans, N., Smith, F. (2011). Disability weights for suicidal thoughts and non-fatal suicide attempts. Journal of Affective Disorders, 134, 341-347. Van Spijker, B.A.J., Werner-Seidler, A., Batterham, P.J., Mackinnon, A., Calear, A.L., Gosling, J.A., Reynolds, J., Kerkhof, AJ.M.F., Solomon, D., Shand, F., Christensen., H. (2018). Effectiveness of a Web-Based Self-Help Program for Suicidal Thinking in an Australian Community Sample: Randomized Controlled Trial. J Med Internet Res,20(2):e15. http://dx.doi.org/10.2196/jmir.8595 310 ― References

Van Winkle, N.W., May, P.A. (1986). Native American suicide in New Mexico, 1959-1979. Human Organization,45,296-309. Van’t Hof E, Cuijpers P, Stein DJ. (2009) Self-help and internet-guided interventions in depression and anxiety disorders: a systematic review of meta-analyses. CNS Spectr,14,34–40. Ventriglio, A., Bhugra, D. (2018). Micro-identities and Acculturation in Migrants. In D. Moussaoui et al. (eds.), Mental Health and Illness in Migration, Mental Health and Illness Worldwide, https:// doi.org/10.1007/978-981-10-0750-7_4-1 Vijayakumar, L., John, S., Pirkis, P., Whiteford, H. (2005). Suicide in developing countries: Risk factors. Crisis, 26,112-119. Wagenmakers, E-J., Love, J., Maarsman, M., Jamil, T., Ly, A., Verhagen, J., Selker, R., Gronau, Q.F., Dropmann, D., Boutin, B., Meerhof, F., Knight, P., Raj, A., van Kesteren, E-J., van Doorn, J., Smira, M., Epskamp, S., Etz, A., Matzke, D., de Jong, T., van den Bergh, D., Sarafoglou, A., Steingroever, H., Derks, K., Rouder, J.N., Morey, R. (2018a). Bayesian inference for psychology. Part II: Example applications with JASP. Psychon Bull Rev,25, 58–76. http://dx.doi.org/10.3758/s13423-017-1323-7 Wagenmakers, E-J., Marsman, M., Jamil, T., Ly, A., Verhagen, J., Love, J., Selker, R., Gronau, Q.F., Smira, M., Epskamp, S., Matzke, D., Rouder, J.N., Morey, R.D. (2018b). Bayesian inference for psychology. Part I: Theoretical advantages and practical ramifications. Psychon Bull Rev. 25:35–57. http://dx.doi. org/10.3758/s13423-017-1343-3 Walker, R.L., Obasi, E.M., Wingate, L.R., Joiner, T.E. (2008). An empirical investigation of acculturative stress and ethnic identity as moderators for depression and suicidal ideation in college students. Cultural Diversity and Ethnic Minority Psychology,14,75-82. Wang, D., Locke, C., Chonody, J. (2013). Differences by Race in Social Work Students’ Social Distancing, Treatment Options, and Perceptions of Causes of Mental Illness.Journal of Social Service Research, 39, 646-661, http://dx.doi.org/10.1080/01488376.2013.816410. Wassenaar, D. R., Marchiene, B. W., van der Veen, M. S. S., Pillay, A. L. (1999). Women in Cultural Transition: Suicidal Behavior in South African Indian Women. Suicide and Life-Threatening Behavior,28, 82-93. Wenzel, A., Brown, G. K., Beck, A. T. (2009). Cognitive Therapy for Suicidal Patients: Scientific and Clinical Applications. Washington, DC. American Psychological Association. Werlang, B. G., Botega, N. J. (2003). A semi-structured interview for psychological autopsy in suicide cases. Rev Bras Psiquiatr,25, 212-9. Westermeyer, J. (1990). Working with an interpreter in psychiatric assessment and treatment. The Journal of Nervous and Mental Disease,178,745-749. White, J.B. (1997). Turks in new Germany. American Anthropologist,99,754-769. Whiteside, U., Lungu, A., Richards, J., Simon, G. E., Clingan, S., Siler, J., Snyder, L., Ludman, E. (2014). Designing, messaging to engage patients in an online suicide prevention intervention: survey results from patients with current suicidal ideation. J. Med. Internet Res, 2,1-8. World Health Organisation (2017). Depression and Other Common Mental Disorders: Global Health Estimates. Geneva: World Health Organization. Retrieved February 19, 2016 from https://www.who. int/mental_health/management/depression/prevalence_global_health_estimates/en/ World Health Organisation (2012) Global health observatory data repository. Retrieved February 19, 2016 from http://apps.who.int/gho/data/node.main.MHSUICID?lang=en. World Health Organisation (2001). The world health report 2001. Mental Health: New understanding, new hope. Retrieved on 12th June 2018, from https://www.who.int/whr/2001/en/whr01_en.pdf?ua=1 Williams J.M.G., Swales M. (2004). The use of mindfulness-based approaches for suicidal patients. Arch Suicide Res 8: 315–29. References ― 311

Williams, C., Martinez, R. (2008). Increasing access to CBT: Stepped care and CBT self-help models in practice. Behavioural and Cognitive Psychotherapy,36,675-683. WMR (2010). Health of Migrants: The Way Forward-Report of a Global Consultation. WHO Press, World Health Organization. World Health Organization (WHO) Global Health Observatory data repository. Suicide rate estimates, age-standardized estimates by country. http://apps.who.int/gho/data/node.main. MHSUICIDEASDR?lang=en. Accessed 17 September 2018. Yalaki Z., Tasar M.A., Yalcin N., Dallar Y. (2011). Cocuk ve genclik dönemindeki özkiyim girisimlerinin degerlendirilmesi [The evaluation of the suicide attempts in childhood and adolescence]. Ege Tip Dergisi,50, 125–128. Yasan, A., Danis, R., Tamam, L., Ozmen, S., Ozkan, S. (2008). Socio-Cultural Features and Sex Profile of the Individuals with Serious Suicide Attempts in Southeastern Turkey: A One-Year Survey. Suicide and Life-Threatening Behavior,38, 467-480. Yektaş, Ç., Bildik, T., Özbaran, N. B., Köse, S., Erermiş, S. (2014). Depresif kiz ergenlerde intihar davranişinin klinik özellikleri [Clinical characteristics of suicidal behavior in depressive adolescent girls. Çocuk ve Gençlik Ruh Sağliği Dergisi,21,33-44. Yılmaz, A. E.,Gencoz, T.,Wells, A. (2008). Psychometric characteristics of the Penn State Worry Questionnaire and Metacognitions Questionnaire-30 and Metacognitive Predictors of Worry and Obsessive Compulsive Symptoms in a Turkish sample. Clinical Psychology and Psychotherapy, 15, 424-439. Yılmaz, T. A., Riecher-Rossler, A. (2008). Suizidversuche in der ersten und zweiten generation der immigrantInnen aus der Turkei [Suicide attempts among first and second generation Turkish migrants] Neuropsychiatrie,22,261-267. Yılmaz, T. A., Riecher-Rossler, A. (2012). Attempted suicide in immigrants from Turkey: A comparison with Swiss suicide attempters. Psychopathology,45,366-373. Zeyrek, E.Y., Gençöz, F., Bergman, J., Lester, D. (2009) Suicidality, problem solving skills, attachment style and hopelessness in Turkish students. Death studies, 33, 815-827. Zigmond A.S., Snaith R.P. (1983) The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand, 67,361-370.

R

a

APPENDICES 314 ― Appendices Methods Methods Use of modelling of Use using case(i.e. examples) to explain usershow can do the exercises; Some mentalization techniques have been added to the “worry time’’ exercise. Information: the methods the of is intervention according tailored cultural to Turkish values. Goals Goals None Information: the goals the of modules are in line with the cultural values. Context Context Use of Turkish Turkish of Use Tarik names (e.g. and Cemil) and context in case studies suicidal of thinking. Information: the context of the and intervention exercises are tailored in the of light previous research with Turkish communities (Eylem et al., 2016) ” and and e.g. ’’ e.g. it it ’’ e.g. religiosity interdependence independence Cultural Adaptation Components Concepts Concepts “It is normal to think“It deathof in Islam but thinking suicide of is different”(religiosity), feeling afraid of letting family down (interdependence) “ interdependence Use of of Use Use of theUse concepts “ is their responsibility the modules to follow (independence) and guidance is ifavailable further assistance is needed (interdependence). to Metaphors Metaphors “don`t give me fish, teach “don`t me how catch I can fish” Use of the of metaphors: Use “KiymaCanina-Don`t crush lifeyour energy’’ instead to name the ‘’suicide’’ of intervention; None participate actively doing by the exercises and homework. Linguistic Adaptations Information: The modules are available in 3 languages (Dutch, Turkish) English, and the relevance and understandability of the has content been checked native by speakers. Linguistic adaptations Reducing literacy, rephrasing the sentences. Introduction Module 1 Thinking about Suicide Intervention Appendix A Cultural Diagram Adaptation 7 Chapter Appendices ― 315 Methods Use of visualizationUse happy picture(e.g. sad you, of picture you) of instead feeling of thermometer; some mentalization techniques have been added to the still“sit and do nothing’’ exercise and the name has been replaced stillas “sit and reflect back’’; the information and exercises about self-harm been have removed. Goals Respecting users` users` Respecting concern possible sharingabout their feelings with someone by them encouraging to imagine themselves sharing their feelings with someone in “an in experiment sharing your feelings” Context Use of Turkish Turkish of Use names and context in case studies. Concepts Use of theUse concept honour) (i.e. “namus” in the case study e.g. is experiencingFatma crisis after has she an argument with her parents being not about allowed over in to stay place.her friends`

A Metaphors Use of metaphors to explain metaphors of Use the crisis “wanting to bang headyour from wall to “feelingwall’’, like your head is going to explode” Linguistic adaptations Reducing literacy, rephrasing the sentences Module 2 Dealing with Feelings and Crises 316 ― Appendices Methods Adding some mentalization techniques to the “feelings of guilt exercise’’ sometimes (e.g. we confuse our thoughts with facts and feel guilty because our of predictions) Goals None Context Use of Turkish Turkish of Use names and context in the case studies; rephrasing automatic thoughts according to the cultural knowledge of such thoughts context in Turkish failed I have my e.g. I am a black family, spot family`s in my I have forehead (i.e. humiliated my family bringing by bad reputation to them) Concepts Use of cultural of Use sadness of expression moral bozuklugu) (e.g. in exercises Metaphors None Linguistic adaptations Reducing literacy, sentences rephrasing theoretical explaining background. Module 3 Automatic Thoughts Appendices ― 317 Methods Perhaps not usingPerhaps not “metaphors the and exaggeration exercise” as metaphors and exaggerations are embedded within culture! Turkish Goals None Context Use of Turkish Turkish of Use names and context in the case studies; giving examples casesfrom Turkish to explain the patterns thinking an example fore.g. thinking negative style-“Cemil feels upset his by friend’s behaviour when they come across supermarket. the at think might He inEveryone the community knows about my daughter’s break-up with her fiancé.No-one will ever respect again. me Why I even do stay thison earth? If he reacts like this, Cemil wellmight end up with some thoughts of suicide.” Concepts Use of concepts-Use namus in the “honour”, examples case

A Metaphors Use of the of and “black white Use in filter in ohotography” order to explain thinking patterns Thinking e.g. particular like are patterns type filters of as black (such limiting filter) & white our perception with only one perspective Linguistic adaptations Reducing literacy, thinking rephrasing patterns according to the cultural knowledge of Turkish populations. Module 4 Thinking Habits 318 ― Appendices Methods Methods Use of modeling of Use in exercises about questioning the automatic thoughts counter finding and thoughts. None Goals Goals The “Does goal of life any have meaning?” exercise has been changed life my worth to “Is living?” exercise. The former had a very general the but goal latter is specific more and is in it accord with the cultural knowledge None counter Some No one No Context Context people do not respect me again’ will ever respectwill me again’, thought- ‘ Use of Turkish Turkish of Use names and context rephrasing whilst automatic thoughts and counter thoughts automatic e.g. thought- ‘ Turkish of Use names and context re-phrasing whilst the case examples Concepts Concepts Use of cultural of Use concepts-namus, man’s whilst pride (honour) rephrasing automatic thoughts cultural of Use concepts-man`s pride (honour) whilst automatic changing thoughts to counter in thethoughts case examples. Metaphors Metaphors None None Linguistic adaptations Linguistic adaptations Reducing literacy, sentences rephrasing theoretical explaining background. Reducing literacy, sentences rephrasing theoretical explaining background. Module 5 Thinking Differently Module 6 Dealing with Future Setbacks Appendices ― 319

Appendix B Chapter 7

Table 5. Model Comparisons Between the Null and the Alternative Models for the Study Variables among Those with Severe Suicidal Ideation Scores on BSS (N=7) Variable Model BF10 Ratio BSS (T0 vs T3) Null model 1.00 a Time 2.567.569 1.00 Group 0.60 0.23 Time + Group 1.840.740 0.33 Time+Group+Time*Group 1.140.252 0.44 BSS (T1 vs T2) Null Model 1.00 a Time 2.688.888 1.00b Group 0.60 0.37 Time+Group 1.889.473 0.71 Time+Group+Time*Group 1.155.267 0.43 BDI (T0 vs T3) Null Model 1.00 a Time 5.289 1.00b Group 1.105 0.21 Time+Group 9.476 1.79c Time+Group+Time*Group 5.161 0.97 BHI (T0 vs T3) Null Model 1.00 a Time 6.943 1.00b Group 0.933 0.13 Time+Group 10.512 1.51c Time+Group+Time*Group 6.765 0.97 PSWQ (T0 vs T3) Null Model 1.00 a Time 0.831 1.00b Group 0.755 0.91 Time+Group 0.691 0.83 Time+Group+Time*Group 0.365 0.48 A EQ5SD (T0 vs T3) Null Model 1.00 a Time 3.490 1.00b Group 0.785 0.24 Time+Group 3.387 0.97 Time+Group+Time*Group 8.176 2.34c Note. T0: Baseline (pre-test); T1: Bi-weekly measures of BSS at week 2; T2: Bi-weekly measures of BSS at week 4; T3: post-test; BF10: Bayesian factor grading the intensity of the evidence supporting the alternative model against the null model Ratio: This column represents the ratio (the likelihood) of the effect of time against the group, time and group and the interaction models. The time model is denominator. The BF10 of each model has been divided by the BF10 of the time model in order to calculate the ratio of each model when compared with the time model.a: The ratio for the null model was irrelevant b:The ratio of time against time is always 1 c: When the ratio is between 0 and 1, there is a weak evidence supporting the alternative model against the time model. When it is greater than 1, that means there is a stronger evidence supporting the alternative model against the time model. 320 ― Appendices

Appendix C Interaction Plots of Outcomes Chapter 7

Figure 3. The interaction plot between group status and time (pre and post-test) indicating changes in BDI (depression) scores in intervention group when compared with the wait-list control group

Figure 4. The interaction plot between group status and time (pre and post-test) indicating changes in BHI (Hopelessness) scores in intervention group when compared with the wait-list control group Appendices ― 321

Figure 5. The interaction plot between group status and time (pre and post-test) indicating changes in PSWQ (Worrying) scores in intervention group when compared with the wait-list control group

A

Figure 6. The interaction plot between group status and time (pre and post-test) indicating changes in EQ5SD (Quality of Life) scores in intervention group when compared with the wait-list control group Turkish migrant populations are at elevated risk for suicidal behaviours compared to their native counterparts in Europe. Additionally, there are similarities in disproportionate rates and characteristics of suicidal behaviours for instance, between Turkish migrant women in Europe and Turkish women in Turkey. Several socio-cultural factors have been proposed to explain the heightened risk such as acculturation di‡ iculties in host countries, the cultural meaning attributed to suicide and the cultural continuity of the characteristics and risk factors through and/or regardless of migration. In spite of the elevated risk, it is known that they do not often receive available psychological interventions mainly due to social (e.g. barriers related to stigma) or organisational (e.g. waiting times and service availability) reasons. In recent years, e-mental health has been introduced to the mental health care as a potential solution for the disparity between the need for services and access to them. However, their e‡ ectiveness is currently lacking for ethnic minorities in general and for Turkish migrants specifically. Thus, this thesis presents various qualitative and quantitative studies with the intention to further the current understanding of suicidality among Turkish migrants. It also presents the results of a feasibility and pilot trial of a culturally adapted e-mental health intervention in reducing suicidal ideation among Turkish migrants in the Netherlands and the UK.