The Yazidis are a small ethnic-religious minority group which is spread over several Middle Eastern countries including Iraq, Syria, Armenia, Georgia and Turkey (Maisel, 2008). They speak Kurdish and most of them consider themselves ethnic Kurds, although some communities prefer identifying themselves as Yazidis with their own ethnic and religious identity (Kreyenbroek and Omarkhali, 2016). There are about 800,000–1,000,000 Yazidis worldwide (Cetorelli et al., 2017), the largest group of about 400,000 Yazidis lives in Northern Iraq in the area of Mount Sinjar in Nineveh governorate (Maisel, 2008). They practice one of the oldest religions, notably Yazidism, which contains elements of Islam, Christianity, Judaism and Zoroastrianism (Cetorelli et al., 2017) and is passed on orally (Kizilhan, 2014). Many derogative prejudices exist against the Yazidi community, one of them being the wrong accusation of engaging in “devil-worshipping” (Maisel, 2008). For this reason, they are considered as “Devil Worshippers” and not regarded as “Followers of the Book” by radical Muslims (Kizilhan, 2017a). Yazidis have been victims of persecution, forcible conversion to Islam and systematic murder for many centuries. In the last 800 years, 74 genocides against this minority group have been perpetrated (Kizilhan, 2017a). The Yazidis have been particularly affected by massacres due to their religion and their secluded settlement areas in the Sinjar Mountains (Kizilhan et al., 2020a) and the lack of protection (Maisel, 2008). For these reasons, it can be argued that the group as a whole has been exposed to a “genocidal environment” with far reaching, also trans-generational impact (Kizilhan et al., 2022), while other groups in the region were not exposed to transgenerational, multiple genocides.

In August 2014, ISIS attacked the Sinjar region of Northern Iraq (Independent International Commission of Inquiry on the Syrian Arab Republic, 2016). The Yazidis were forced to seek refuge on Mount Sinjar (Independent International Commission of Inquiry on the Syrian Arab Republic, 2016). More than 3,000 of them were killed and almost 7,000 kidnapped (Cetorelli et al., 2017). Thousands of Yazidis were tortured, suffered inhuman and degrading treatment, were held hostage, or were forced to convert to Islam. Young boys were separated from their families and placed with ISIS fighters and women and young girls were raped and sold on Arab markets (Cetorelli et al., 2017; Kizilhan, 2017a). Many of those who managed to flee remained trapped in tremendous circumstances without water, food, shelter, and enduring extreme temperatures and were killed or kidnapped if they could not flee in time. They were evacuated between the 9^th and 13^th of August and finally managed to flee through Syria into the KRI (Yazda, 2017). The majority was displaced in camps in the KRI, others settled in camps in Syria or Turkey (Daloglu, 2014; Sidky and Rummery, 2014). While some Yazidis returned to their home villages (Spencer, 2015), many of them found asylum in foreign countries including Germany or Turkey (Tekin et al., 2016; Kizilhan, 2017a). But the biggest part still lives as internally displaced people (IDP) in so-called IDP camps or informal settlements (Yazda, 2017). Most of the severely traumatized Yazidi women live in one of the 24 refugee camps near Dohuk and Zakho in the KRI. Each camp consists of up to 28,000 IDP (Kizilhan and Noll-Hussong, 2017). There are an estimated 360,000 Yazidis who have lived in camps for IDP in the KRI (Yazda, 2017). Not only did the Yazidis experience their own individual trauma, but they also experiences a collective and transgenerational trauma which is passed on to the next generations (Kizilhan and Wenzel, 2020).

The situation in these IDP camps in Iraq and also, in general, the situation in refugee camps is often extremely precarious (Cetorelli et al., 2017). A systematic review of psychiatric disorders among forcibly displaced people in conflict settings demonstrated prevalence rates mental disorders much higher even than in more general post-conflict and war-zone populations, for PTSD up to almost 90% as well as for depression and anxiety disorders of up to 80% (Morina et al., 2018). Surveys show that remoteness, poor connection and poverty in the camps lead to a lack of supply and care (Yazda, 2017; Jäger et al., 2019).

Recently, a new set of problems arose due to the current COVID-19 pandemic which has exacerbated the prior psychological stress. Social distancing is almost impossible in the overcrowded IDP camps and hygiene practices are difficult to apply (Raju and Ayeb-Karlsson, 2020). A study by Kizilhan and Noll-Hussong (2020) compared the prevalence rates of mental disorders before COVID-19 and shortly after COVID-19 outbreak of Yazidis living in IDP camps in the KRI. It was found that PTSD rates increased significantly to up to 58% among women and up to 47% among men. Also depression, anxiety, dissociation, somatoform disorders and suicidal ideas increased (Kizilhan and Noll-Hussong, 2020).

Prevalence rates of mental disorders are estimated to be even higher in survivors of rape, military action, captivity, internment for ethnic or political reasons, or genocide (American Psychiatric Association, 2013). For example, research conducted in Rwanda (Schaal and Elbert, 2006) and Bosnia (Weine et al., 1995) shows that genocidal atrocities have severe long-term effects on the people concerned. Forty-four percent of Rwandan adolescents suffered from PTSD 10 years after genocide, but the percentage of the adolescents presenting some of the PTSD-related symptoms is even higher (Schaal and Elbert, 2006). Among the Bosnian refugees, up to 65% suffered from PTSD and 35% from depressive disorders a few months after their resettlement in the United States (Weine et al., 1995).

During genocide, sexual violence against women is a serious problem and rape is used frequently as war tactic because of its devastating physical and psychosocial consequences on the individuals and also on the whole community (Ruby Reid-Cunningham, 2008). When rape is committed on a mass scale, consequences are even more serious (Ruby Reid-Cunningham, 2008). This reality was illustrated by different studies conducted in Bosnia-Herzegovina, Rwanda, Sudan, Uganda and Congo where rape and extreme sexual violence were carried out against women (Amnesty International, 2004; Lončar et al., 2006; Cohen et al., 2009; Amone-P’Olak et al., 2016; Albutt et al., 2017).

Whether people develop a mental disorder after a potential traumatic event (PTE) depends on various risk factors (Kizilhan et al., 2020a). First and foremost, higher exposure to traumatic experiences (Steel et al., 2009) and post-migration stress (Bogic et al., 2015) seem to play a particularly important role. Other risk factors include the severity and type of PTEs (Darves-Bornoz et al., 2008), pre-existing mental disorders such as depression and PTSD (Stander et al., 2014) and individual characteristics (Farhood et al., 2016), feeling of guilt (Kubany et al., 1995), and the time interval (Steel et al., 2009) since the PTE. In war-affected populations, torture (Steel et al., 2009), captivity (Farhood et al., 2010), sexual attacks (Wright et al., 2017), experiencing violent attacks (Johansen et al., 2013), and the death of a significant, close person (Tekin et al., 2016) were correlated with PTSD and depression. In a sample of Yazidi women, Kizilhan et al. (2020a,b) found that the number of PTEs experienced and the duration as well as the frequency of captivity predicted comorbid PTSD and depression.

Consequently, the psychosocial stress in these IDP camps is extremely high and many Yazidis suffer from serious mental health problems (Haroz et al., 2018). It was found that 42.9% of Yazidi refugees in Turkish refugee camps suffer from PTSD and 39.5% of major depression (Tekin et al., 2016). In a study with 416 Yazidi women and girls (65 of whom had survived sexual enslavement) in IDP camps in the KRI, more than 80% and almost all survivors who had experienced enslavement fulfilled criteria for PTSD (Ibrahim et al., 2018). It has been shown that formerly enslaved Yazidi females suffer from higher prevalence of mental stress (97.1%), PTSD (90.6%), suicidal ideation (38.1%), depression (36.7%), and general anxiety symptoms (37.4%) (Taha and Slewa-Younan, 2020). Among those who had been sexually abused, very high prevalence rates of mental stress of 50.9% for complex PTSD and 20% for PTSD were found (Hoffman et al., 2018). These high rates, particularly for complex PTSD, emphasise the unique nature of trauma caused by the combination of captivity and sexual slavery. An alternative explanation for the occurrence of complex PTSD might be the stress and poor conditions in the refugee camps after the captivity by ISIS (Hoffman et al., 2018).

Subsequent life events, physical and psychological support, and the role of religion and culture have been shown as important protective factors (Murthy and Lakshminarayana, 2006). Furthermore, daily stressors, such as material deprivation and loss, interparental conflict, abuse, concerns about perceived safety and basic needs, can mediate the relationship between war exposure and mental health (Fernando et al., 2010; Rasmussen et al., 2010).

Suicidal behavior is also an important issue in post-conflict settings (Haroz et al., 2018; Canetto, 2021). Years after a war ended, suicide rates seem to have risen in post-conflict settings (Amowitz et al., 2004; Eytan et al., 2015; Mugisha et al., 2016). For example, surveys conducted in Kosovo 8 years after the war and in Rwanda 14 years after the genocide show that 11% of the Kosovo participants and almost 70% of the Rwanda sample reported that they had already felt suicidal ideation or considered hurting themselves (Eytan et al., 2015). Additionally, Wenzel et al. (2009) reported suicidal ideation to be associated with unemployment, and with high depression and PTSD scores. In post-conflict Northern Uganda, 12.1% stated suicidal ideation and 6.2% suicidal attempts (Mugisha et al., 2016). Suicidality was found to be higher in this Ugandan population when the people suffered from depression and PTSD (Flory and Yehuda, 2015). In Southern Iraq, high attempted-suicide rates of 5 to 7% were found and even higher rates of suicidal ideations of over 20% (Amowitz et al., 2004).

This study investigated the relationship between psychological stress, particularly among Yazidi members living in refugee camps in Iraq and Yazidi members and Non-Yazidi who lived in a more or less secure Kurdistan Region of Iraq during the crimes perpetrated by ISIS.

We excluded those community members with a reported clinical diagnosis of severe mental illness such as a psychosis, which might interfere with informed consent or understanding of the questions in the study instruments. Furthermore, two participants were excluded from the data analysis because they could not be assigned to one of the three groups in terms of their living situation or religion.

The aim of this study was to investigate the relationship between psychological stress among members of the Yazidi community 7 years after the genocide perpetrated by ISIS in Iraq in 2014. Three groups were analyzed with regard to their physical and mental health: (1) Yazidis who experienced violence at the hands of ISIS and now live in refugee camps in the KRI, (2) Yazidis who were not attacked by ISIS and do not live in refugee camps and (3) a control group of Muslim citizens who live in the same area but tend to experience comparatively lower levels of discrimination. Psychological stress and suicidality were measured and compared in the three different groups. Furthermore, the relationship between socioeconomic factors and suicidality was examined.

Significant differences in psychological stress (general physical health, general mental health, level of depression, level of anxiety, subscales of PTSD such as intrusion, avoidance and hyperarousal) were found between the three groups. Post-hoc comparisons (Dunn-Bonferroni tests) for the Kruskal–Wallis test show that the Yazidis exposed violence by ISIS reached higher scores for perceived burdensomeness and thwarted belongingness than Yazidis who were not attacked by ISIS. Interestingly, the scores for perceived burdensomeness of attacked Yazidis and Muslim citizens did not differ significantly, and Muslim citizens reached even higher scores for thwarted belongingness than the group of the attacked Yazidis.

The Yazidis exposed violence also showed lower scores for general physical health, general mental health, and higher scores for level of depression, level of anxiety, subscales of PTSD such as intrusion, avoidance and hyperarousal in comparison with the Yazidis not exposed violence. However, Yazidis exposed violence by ISIS did not always have a significantly higher (or lower) score than the non-Yazidis. The Yazidis not exposed violence reached lower scores for general physical health, higher scores for intrusion, avoidance and hyperarousal. However, this was not the case for general mental health, level of depression and level of anxiety, as the Yazidis exposed violence always had higher psychological stress scores (or lower scores for general physical and mental health) than the Yazidis not exposed violence, but not always higher scores (or lower scores for general physical and mental health) than non-Yazidis.

Further analyses were conducted to measure the relationship between socioeconomic variables – such as age, years of education, gender, employment, physical violence, loss of important people, flight experience, number of siblings - and suicidality. There was a small negative correlation between the duration of education and perceived burdensomeness, which was statistically significant. The fewer years of education a participant had, the higher was the score for perceived burdensomeness. Moreover, it was found that employment significantly predicts thwarted belongingness, as does physical violence and loss of important people.

As the results show, the Yazidis who were attacked by ISIS always had higher scores for suicidality and psychological stress compared to the Yazidis not exposed violence, but not always compared to the non-Yazidis. The score differences were especially high for PTSD. These results are to be expected and understandable to a large extent, as Yazidis exposed violence has experienced by far most stressors compared to Yazidis not exposed violence and non-Yazidis. Many of them have experienced sexual or physical violence, had to flee, been in ISIS captivity, lost important people in their lives, generally have low levels of education, and are mostly unemployed. All the mentioned factors could explain why these individuals are at higher risk for psychological stress and suicidality. Furthermore, the higher risk might be increased additionally by difficult living conditions in refugee camps (Yazda, 2017), the current COVID-19 pandemic (Kizilhan and Noll-Hussong, 2020) and other daily stressors (Fernando et al., 2010; Rasmussen et al., 2010).

However, it is interesting to observe that non-Yazidis scored the highest on thwarted belongingness and the same as Yazidis exposed violence on perceived burdensomeness. Participants of non-Yazidis are Kurdish Sunni Muslims and, themselves, represent an ethnic minority in Iraq (Mohamed, 2014). They make up only 16% of the Iraqi population, whereas the majority of 78% consists of Arabs. Whilst the majority of Iraqi Kurds are Sunni Muslims, 62% of Iraqi Arabs are Shia Muslims and 30% Sunni Muslims (Mohamed, 2014). Although they are not a minority in Iraqi Kurdistan, they are still a religious and ethnical minority in relation to the whole population in the country. There has been a lot of tension and fighting between Kurds, Arab Shias and Arab Sunnis (Szanto, 2020). Especially, the hostility between Kurds and Arab Shias has increased over the last years due to the opposition of the Shia government to the referendum on Kurdish independence and the Kurds blaming the Shia for the alienation of Sunni Arabs and the rise of ISIS. Furthermore, the genocide against the Kurds forced many of them to flee and a lot took on new groups of religion which are sometimes anti-Shia (Szanto, 2020). This has further increased the prior split between the two groups. These factors could explain the high scores of the non-Yazidis for thwarted belongingness. It is reasonable that after experiencing such persecution and hostility, non-Yazidis is still marginalized and does not feel that they belong properly to the society they are living in, nor do they feel valued and appreciated within the country. Moreover, all these conflicts could also have an influence on the perceived burdensomeness which would explain why non-Yazidis did not show significantly lower scores on this variable.

All participants, both Yazidi and Muslims, scored low on perceived burdensomeness. One explanation could be that people from this culture generally think in a more collective way in which personal feelings and emotions are not considered as so important (Kizilhan, 2017b).

Moreover, there is a different understanding of mental illness, whereby it is seen as a consequence of fate and not really as an interaction of various factors (Kizilhan, 2017a). This fact can lead to a stigmatization of mental illness problems. All these aspects can explain why the participants might not have been able to adequately express their feelings and emotions about their perceived burdensomeness, as the collective society is regarded to be more important than the individual well-being. Moreover, suicidality might also be a taboo subject in this society (e.g., Jäger et al., 2019), thereby further constraining individuals and thus perpetuating the vicious circle.

Physical violence is often associated with trauma and PTSD (Resnick et al., 1993; Kessler et al., 1995), which can lead to a disordered worldview. Similarly, the fact of losing important people in life may damage the sense of belongingness as well as the experience of being valued. This is due to the fact that they substantially contribute to the feeling of fitting into the broader social context.

We found a direct relationship between physical violence, loss of important people, flight experience, duration of education, gender and perceived burdensomeness. These are all factors that increase the perceived burdensomeness at different levels. Being a woman in the KRI is associated with many obstacles, less freedoms and more restrictions (Joly and Bakawan, 2016). There is a deeply-established gender inequality between men and women in Iraq which is also supported by the Iraqi law (Joly and Bakawan, 2016). Patriarchal norms and traditions further increase men’s domination over women (Joly and Bakawan, 2016) and the prevalence of intimate partner violence is very high and has significant consequences in terms of serious injuries (Al-Atrushi et al., 2013). Women are seen as property of the family and often represent the symbol of a family’s honor and dignity in terms of purity, virginity and modesty (Laizer, 1991). These circumstances could lead to a culture in which women perceive themselves increasingly as a burden to others by feeling that they are not important enough and therefore represent a burden to society. This, in turn, would constitute an additional challenge for their family or society.

There are countries in which women have higher suicide rates than men and therefore represent an exception to this gender paradox. Examples include China, Pakistan, Bangladesh, Indonesia, Nepal, and also Iraq. It has been shown in several previous findings that there is an association between violence and suicidality (Devries et al., 2011), intimate, physical and emotional partner violence and suicidal behavior (Weaver et al., 2007), and also between rape and suicidality. In Kurdistan, women are often victims of intimate partner violence (Al-Atrushi et al., 2013). For this reasons, physical violence could be a mediator between gender and suicidality in the context of women in the KRI. It may lead to new insights to carry out further mediating analyses to examine the link between gender, physical violence and suicidality.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/supplementary material.

The studies involving human participants were reviewed and approved by University of Duhok IPP. The patients/participants provided their written informed consent to participate in this study.

JK planned the study and conducted the study with support of his research team. LS analyzed the data and wrote the draft manuscript with support from JK. LS was supervised by TB. TW contributed by revising the manuscript and reanalyze some of the data. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.