Background

Suicide and attempted suicide affect millions of people worldwide and are major tragedies at the family, community and personal level. Suicide refers to “death resulting from intentional self-injurious behavior, associated with any intent to die as a result of the behavior”, while attempted suicide refers to “a potentially self-injurious act committed with at least some intent to die, as a result of the act.” [1] Globally, it is well documented that in most countries, women attempt suicide more often than men, although men die by suicide more often than women, partially because they use more lethal methods [2, 3]. An exception to this longstanding gender pattern in suicide deaths is the high number of women who die by suicide in certain Muslim-majority countries in the Middle East and parts of Asia [2, 4–8]. This variation is even more striking when stratified by rural (generally more traditional and somewhat more religious) regions versus urban (generally less traditional and somewhat more secular) regions. For instance, studies have suggested that young Pakistani women living in rural and/or remote areas [6], women in rural Iran [7] and Turkish women and girls in rural areas of Eastern Turkey [8, 9] have a higher risk of suicide than males in these regions. Therefore, we choose one of these Muslim/majority female groups in relation to suicidal behaviour, that is, Turkish females, as the topic for investigation for this paper. For Turkish females, 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 [10]. This is why the suicidal behaviours of Turkish females who live in Turkey as well as Turkish female immigrants to Europe need further investigation [11].

The epidemiology of suicidal behaviour of Turkish women (residing in Turkey) shows that the suicide rate among Turkish women aged 15–24 in Eastern Turkey is twice as high as their male counterparts since 2000 [9]. These finding were recently replicated by an annual report on suicide statistics drawn from Turkey’s general population [12]. Similarly, hospital registration data in Southeast Turkey has shown that attempted suicide rates were 4.15 times higher among young women 15–24 years old who presented at emergency departments compared to men of the same age as well as older men and women [11, 13].

Focussing on the epidemiology of suicidal behaviour of Turkish female immigrants in Europe, hospital registration studies show that in the Netherlands [14], Germany [15] and Switzerland [16], the risk of attempted suicide was at least 2.5 times higher among Turkish immigrant women aged 14–25 who presented at emergency departments compared to Turkish immigrant men of the same age and women in the host country. Furthermore, 10–17-year-old girls and young women of Turkish descent had an almost twofold risk of suicide compared to ethnic majority women in Germany (relative risk = 1.79; 95% CI = 1.41–2.27) [17].

Regarding the demography of migrants in Europe, it is striking that immigrants from Turkey are among the largest ethnic minority populations in multiple western European countries. Most came as contracted foreign labourers in the 1960s and 1970s from predominantly rural and traditional areas in Turkey [18]. The countries that received the most Turkish immigrants were Germany, Switzerland, Austria and the Netherlands, which have Turkish immigrant populations of 5%, 5%, 3% and 3%, respectively [18]. People who migrate from Turkey to Europe are predominantly Muslim (approximately 95%), and most are Sunni [18]. During the 1960s and 1970s, socio-economic deprivation, especially in the rural areas of Eastern Turkey, pushed migrants to seek employment in Western Europe, which eventually led to family reunification and cross-national marriages. More recently, migration from Turkey has been triggered by political instabilities [19, 20].

The elevated risk of suicidal behaviours among young Turkish women in Turkey and female Turkish immigrants in Europe raises important questions that we will address in this review: 1) what characteristics and precipitating factors contribute to the elevated risk of suicidal behaviour among Turkish females? 2) To what extent is there similarity in the characteristics and precipitating factors contributing to suicidal behaviour of Turkish women residing in Turkey versus in Europe? We use the term ‘precipitating factors’ to describe perceived causes or contributors to the manifestation of suicidal behaviour, and these perceptions can be either participants’ views or researchers views (depending on the methodology used in studies).

Precipitants and characteristics of suicidal behaviours among Turkish women

Mental illness is often listed as a critical factor in suicidal behaviours in individuals [21]. In a study from Southern Turkey, depression was three times more likely to be reported among women than men who attempted suicide (72% versus 27%; [22]). Similarly, depression was more often reported among Turkish female immigrants who attempted suicide in Germany compared to German majority women (51% versus 34%; [23]). Although depression and other mental illnesses have been linked to suicide, several authors have argued that they are not the ultimate cause of suicidal behaviour among women of certain cultural backgrounds. Indeed, these authors suggested that mental illness and suicidal behaviours among women in the Middle East are at least partially a consequence of aggression or violence against women rather than an independent factor or psychopathology [2]. A similar argument has been put forward for immigrant women from Turkey and South Asia in Europe who have demonstrated suicidal behaviour [24, 25].

Turkey’s socio-cultural landscape is characterised by a variety of ethnicities, and religious convictions and relatedly, a diversity of values and norms. Ethnic Turks and Kurds together comprise the largest ethnic groups in Turkey, but the country also hosts ethnic populations much smaller in size, for instance Circassian, Bosnian, Albanian, Georgian and Arab Turks [19, 20]. Most Turks are Sunni Muslims (81%), but a minority are Alevis, a group that has a more secular lifestyle than Sunnis [18]. Traditionalism (e.g., strong identification with Islamic ideology and a patriarchal family life) is common in the central and eastern regions, while secular lifestyles and egalitarian views of gender roles are more common in the western regions [26]. The concepts of namus and seref (i.e., honour) are central to how gender roles are expressed according to the patriarchal system of rural Turkey [10, 26]. Namus holds women responsible for their family’s honour by having them maintain their sexual abstinence until they get married and requiring them to demonstrate modesty in their interactions with the opposite sex. Failing to do so can result in the loss of a family’s reputation [26]. In extreme cases, so-called honour killings are committed by male family members and/or male community members against women who are seen to jeopardise their family’s reputation [26, 27].

Namus and seref can play a role in the manifestation of suicidal behaviours by women living communities in traditional rural Turkey [9, 28, 29]. There is also evidence that these phenomena can be transported to Europe upon migration by Turkish people. Studies showed that honour-related violence and aggression were relevant to suicidal behaviours among young Turkish immigrant women in Germany [10], the Netherlands [24]. Honour-related violence was also reported among young women in other Muslim-majority countries in the Middle East [6–10, 30]. Furthermore, empirical findings of the impacts of honour-related violence for suicidal behaviour resonate with Durkheim’s archetype of fatalistic suicide [24, 31, 32]. According to this archetype, suicide is a reaction to the harsh moral demands of the community, which are upheld by force [24]. When strict moral rules are experienced by women as external, limiting, demanding and obtrusive, they may experience hopelessness and, ultimately, suicidal thoughts [8, 31, 32]. The consistency of honour-related violence among Turkish women in Europe and Turkey [24, 25] might be indicative of the continuity of Durkheim’s fatalistic suicide after migration.

Besides the fatalistic suicide model, explanations of the elevated rates of suicidal behaviours among female Turkish immigrants in Europe include difficulties in the migration and acculturation processes [25, 27]. Several studies report that ‘culture conflicts’ cause decreased psychological well-being and suicidality among female Turkish immigrants in Europe [33, 34]. Studies of culture conflicts suggest immigrant women may perceive an unbridgeable gulf between the need to fulfil modern social roles in mainstream society and their abiding the collectivism and traditionalism required by their families and communities [10, 24, 33]. These issues have an impact on suicidal behaviour [24, 25].

Furthermore, socio-economic factors are known to play a role in the mental health and wellbeing of immigrant populations. Turkish immigrants in Europe often suffer from a disadvantaged socio-economic position compared to the majority of the population, including a lack of adequate housing, optimal schooling and access to culturally sensitive mental healthcare. Socio-economic disadvantages are known to predict distress, depression and, ultimately, suicidal behaviour and are therefore relevant to understanding suicidality among Turkish immigrant women [35]. In summary, the intersection of socio-economic factors, honour-related violence, mental illness and culture conflicts is crucial to the understanding of suicidal behaviour among Turkish women [30, 36].

The present study

The heightened risk of suicidal behaviours among Turkish women living in Turkey and Europe is a serious public health problem, and a comprehensive systematic review of this issue is currently lacking. Thus, in this study, we compared and synthesised the empirical evidence of the characteristics and precipitating factors in the suicides and attempted suicides of Turkish women in Turkey and Europe. Specifically, we investigated the relevance of 1) socio-demographic; 2) migration-related (e.g., acculturative stress) and, 3) gender-related factors (e.g., honour-related violence and other forms of violence against women). We choose these three categories as previous research among women and immigrants has shown the relevance of these categories to the emergence of their suicidal behaviour [21, 25, 27]. We also examined the role of psychiatric illness, although we are aware that it might be caused by the aforementioned factors. Since rates of suicidal behaviours and the accompanying precipitating factors may vary depending on degree of traditionalism of the area, we classified the research sites into ‘traditional areas’ and ‘less traditional areas’ (these were assessed by OE and checked by a sociologist from Turkey). We use the term ‘precipitating factors’ when referring to the perceived causes of or important contributors to suicidal behaviour (e.g., different forms of violence) and ‘characteristics’ when referring to the aspects that increase the likelihood of experiencing a suicidal crisis (e.g., psychiatric illness) and/or are not necessarily causal [2, 36]. Some studies distinguished between first-generation versus second-generation immigrant women, terms that respectively refer to whether an individual was born in Turkey and migrated to a host country or was the child of immigrants who moved to the host country.

Methods

Selection and identification of the studies

Search strategy.

We searched the following bibliographic databases: PsycINFO, PubMed, Med Line, Web of Science, Smart Cat, Safety Lit, BASE and Ulakbim. The Turkish database Ulakbim was chosen because it is a comprehensive database that includes articles in Turkish from several Turkish databases, such as Tübitak (i.e. a comprehensive database of scientific publications from variety of disciplines in Turkish, see https://ulakbim.tubitak.gov.tr/en) and Sosyal Bilimler (i.e. a comprehensive database of scientific publications from social sciences such sociology, history, anthropology etc., see https://www.sosyalbilimler.org/). We also checked the reference lists of all the papers we included in our last step in order to identify additional studies [37]. The following terms were used as index terms or free-text words (including synonyms and closely related words): ‘suicide’, ‘deliberate self-harm’ and ‘Turk’. Additional language filters were used to search Turkish, German and Dutch articles within our retrieved records. There is no review protocol for this study. The string for the electronic search strategy was as follows:

1. ‘suicide’
2. ‘deliberate self-harm’
3. 1 or 2
4. ‘Turk’
5. 3 and 4

Inclusion criteria.

The inclusion criteria were 1) articles presenting original empirical data on demographic, psychological, psychiatric or social factors relevant to the suicide or attempted suicide among Turkish girls and women living in Turkey and Europe; 2) articles using hospital samples, crisis-centre samples (for attempted suicide) and/or forensic reports (for suicide); and 3) articles published between 1960 and 2020.

Exclusion criteria.

The exclusion criteria were 1) articles that investigated suicidal ideation only, 2) articles that did not investigate gender differences, 3) articles that only investigated geriatric populations, and 4) articles that only reported standardised mortality rates, prevalence or incidence data.

Quality assessment.

In line with a multidisciplinary approach to examining the characteristics and precipitants of suicidal behaviours [21], multiple research designs were included in the review. The following core criteria were used to analyse the studies (see Table 1): 1) whether specific information about women was assessed and provided (beyond speculation in the Discussion), 2) whether precipitating factors and characteristics of suicide and/or attempted suicide were assessed and reported, 3) whether the articles was published in a peer-reviewed journal, 4) whether the research questions and study design were clear, 5) whether there was a comparison group from the majority ethnic group and/or healthy controls in case-control studies, 6) whether the information on attempted suicide was collected through a retrospective analysis. 7) whether the researchers directly gathered information from the patients and/or relatives (through self-report questionnaires and/or interviews) and 8) whether there was a follow-up assessment.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Quality assessment of the studies.

https://doi.org/10.1371/journal.pone.0253274.t001

Furthermore, the quality of autopsy studies was evaluated by consulting the guidelines for such study designs [64]. The studies that did not meet the first four criteria were excluded (see Fig 1). Two authors (DvB and OE) were involved in the quality assessment, and the discrepancies between the authors’ selections were resolved through discussions.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Prisma flowchart: Flow of studies on suicide and attempted of Turkish women in Turkey and Europe.

Eligible studies were retrieved by the systematic literature search and article selection process, consisting of the following steps: identification of records, screening for inclusion, assessing of studies eligibility, and inclusion of studies into the integrative research review.

https://doi.org/10.1371/journal.pone.0253274.g001

Data extraction.

A customised data extraction form was generated in consultation with the Cochrane Collaborative GRADE approach [37]. This form included but was not limited to countries/regions, population characteristics, sample, place and year of recruitment, methods of data collection, type of suicidal behaviour (suicide or attempted suicide), socio-economic factors, age, gender, precipitating factors, social stressors and psychiatric factors. The titles of all articles were screened, and the abstracts of the articles were checked regarding our inclusion criteria. When no definitive decision could be made based on the abstract, the original articles was reviewed. Two reviewers (DvB and OE) independently completed the extraction form, and a third reviewer (AHM) established consensus across the two forms. After the selection process, we retrieved the full texts of the articles that met the criteria in order to perform our analyses.

Results

After examining a total of 10,189 abstracts, we retrieved 332 full-text papers for further consideration. We excluded 9,857 retrieved papers. A PRISMA flowchart [37] describing the inclusion process is presented in Fig 1. A total of 36 studies were included in the qualitative synthesis; results of the characteristics and precipitating factors are summarised in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Characteristics and precipitating factors of attempted suicide and suicide of Turkish women in Turkey and Europe.

https://doi.org/10.1371/journal.pone.0253274.t002

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). Our analyses of sociodemographic factors showed that 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, and tended to have low socio-economic status. Another remarkable finding was that some studies indicated that religiosity among women who attempted suicide in traditional areas of Turkey was very low [52, 55]. This may suggest that women who are not religious could experience a sense of isolation because they are surrounded by mostly strict believers in traditional areas. The latter argument is supported by recent studies showing that in Muslim countries where the freedom of religion is restricted or religion is compulsory, the protective function of religion could be limited [2, 65].

Our analysis of the role of mental illnesses showed that women often presented symptoms of mental illnesses, such as depression. Moreover, in some studies, mental illness (notably depression) was more commonly reported among women than men or ethnic-majority women (in Europe). However, it is needed to delve deeper into what would cause depressive symptoms in women. This relates to the work of Devries and colleagues, who provided meta-analytical evidence of the predictive role of intimate partner violence on symptoms of depression and attempted suicide among women worldwide [66]. Arguably, the frequency of mental disorders in this current review points to a lack of physical, sexual and emotional safety among women. Impaired autonomy, economic dependence and poverty render women particularly vulnerable to distress, which is a tremendous setback to mental health equity [30, 67]. An interesting finding is that women living in traditional areas of 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 [68]. 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 scientists to medicalise social and political problems, including violence, while the problem of high suicidality should be combatted as a form of social oppression [69, 70].

As for precipitating factors of suicidal behaviours, it was striking that women in the studies had often been 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. Furthermore, there is evidence indicating that violence against women may be labelled as ‘relationship problems’ or ‘family conflict’ [24, 25]. These violations are gender specific and more often a cause for concern for women than men. Cross-culturally, there is also well-documented evidence highlighting that violence against women from their family or spouse is crucial to suicidal behaviours, especially among women of ethnic minority or immigrant backgrounds [67, 71] and that this type of violence triggers suicidal behaviours [67].

One of the important highlights of the current study is the evidence of the restrictions on personal autonomy in the lives of Turkish women in both continents. Women and girls who attempt or die by suicide, particularly if they live in or originate from rural areas of Turkey, seem to be confronted with situations in which their input on strategic choices regarding marriage, divorce, labour market participation and sexuality is infringed upon or denied. Rural areas of Turkey are well known for their practice of conservative gender typing and patriarchal family structures [8, 9, 27, 28]. Although honour-related violence and autonomy restrictions were not always investigated as precipitating factors (28% studies from Europe and 12% from Turkey examined it), it is suspected that the within the ‘domestic violence’ and ‘family problems’ categories, honour played a role as well. A study (not part of our review) of a large community sample of women aged 13–54 (N = 414) in Turkey provided extensive evidence of honour-related violence in traditional rural areas [72]. Accordingly, 55.8% of women under 20 included in this report were married through an arranged marriage, and 24.5% were forced into a marriage. Furthermore, all types of violence were frequently mentioned by these women (99.3% psychological violence, 89.4% physical violence, 91.3% verbal violence, 90.8% economic violence and 63% sexual violence) [72].

Comprehensively viewed, the results of our review could be interpreted within the framework of intersectionality in which the broader structural inequalities in gender roles and expectations, such as self-sacrifice for the sake of honour, and power imbalances among social classes, such as the low socio-economic status of women as well as poverty, collectively impact the suicidal behaviours of Turkish women [2, 5, 25, 29, 55, 69, 73]. Moreover, the consistency of socio-demographic characteristics emphasises that young Turkish women often occupy the lowest social status in highly traditional Turkish communities compared to males and older women and men [10, 72, 74].

Furthermore, the precipitants related to autonomy restrictions also confirm features of Durkheim’s fatalistic suicide model, which suggests that suicide is the result of a lack of control over one’s life, which seems to be dictated by powerful others [31]. In a Durkheimian sense, a fatalistic suicide communicates a sense of alienation in response to harsh moral demands upheld through force [25, 32, 75]. Additionally, drawing on the interpersonal theory of suicide developed by Joiner [76], it could be argued that suicide attempts and suicides are strategies to escape the sense of powerlessness and abuse that evoke feelings of thwarted belongingness [77].

Conclusions

To our knowledge, this is the first systematic literature review comparing the characteristics and precipitating factors associated with suicidal behaviours in Turkish females 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 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.

In summary, the suicidal behaviours of Turkish women are related to a complex interaction of cultural, social, economic and individual factors that vary depending on region and socio-political contexts [2]. Suicidality among Turkish women 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 [25]. These are clear examples of serious human rights violations [2]. There is empirical evidence from the Netherlands indicating that change of cultural beliefs in gender-specific honour codes is possible through psycho-educational training programmes involving ethnic groups that place considerable value on honour [10]. In addition to such programmes, culturally and gender-sensitive services for survivors of domestic violence and early intervention and/or detection of young women at risk are important. See, for example, the awareness campaign End Your Silence Not Your Life, which targets Turkish immigrant women in Berlin [10]. Finally, adequate jurisdictional and political action to prevent human rights violations against women in Europe and Turkey are much needed in order to prevent suicidal behaviours [78, 79].

Supporting information