Suicide and Suicide-related Behavior among Bhutanese Refugees Resettled in the United States

Abstract

Suicidal behavior and death by suicide are significant and pressing problems in the Bhutanese refugee community. Currently, Bhutanese refugees are dying by suicide at a rate nearly two times that of the general U.S. population. Proper identification of risk factors for suicide saves lives; however, if risk is underestimated due to culturally inflexible risk assessments, preventable deaths may continue to occur. In this review, we examine specific cultural factors related to psychopathology and suicide among Bhutanese refugees. To contextualize the current suicide crisis among Bhutanese refugees, we propose an integration of the interpersonal psychological theory of suicide (Joiner, 2005) and the cultural model of suicide (Chu, Goldblum, Floyd, & Bongar, 2010). We provide recommendations for preventing suicide from a systems framework and discuss practical lessons from a preliminary study designed to test a culturally-responsive model of suicide in Bhutanese refugees.

Public Health Significance and Purpose

Globally, the rate of suicidal ideation, planning, and attempts is high at 9.2% (Nock, Borges, Bromet, Alonso, et al., 2008) and the rate of suicide is 10.7 per 100,000 people (World Health Organization, 2017). The global suicide rate is steadily decreasing, down from 13.1 per 100,000 people in 2000 (World Health Organization, 2017). However, in the United States, suicide rates have risen consistently since 1999 (Centers for Disease Control and Prevention, 2016), and suicide is now among the top 10 leading causes of death (National Center for Health Statistics, 2014). Nearly 1.3 million American adults attempt suicide annually with an age-adjusted suicide mortality rate of 13.4 per 100,000 people (National Center for Injury Prevention and Control: Division of Violence Prevention, 2015). There are specific populations who are at particularly high risk for suicide. One such population, Bhutanese refugees, is the focus of this review.

With over 95,449 Bhutanese refugees resettled in the United States since 2008 (Worldwide Refugee Admissions Processing System (WRAPS), 2018), Bhutanese refugees have an age-adjusted suicide rate of 24.4 per 100,000—nearly twice the rate for the general United States population—a substantial public health concern (Ao et al., 2012). Exacerbating this public health crisis is the possibility that contagion effects increase risk among community members (Ao et al., 2012), and under-reporting of suicidal ideation leads to poor detection of risk (Adhikari, Yotebieng, Acharya, & Kirsch, 2015; Ao et al., 2012).

This article will review and consolidate the literature to inform the goal of identifying culturally-sensitive, accurate, and early risk factors for suicidal behavior among Bhutanese refugees. To set the stage, a brief historical overview of the context of the resettlement of Bhutanese refugees in the U.S. is first presented. Prefaced by the definitions of suicide-related terminology we adopted for this paper (e.g., suicide attempt, suicidal ideation, suicidal desire, suicide plan), we then review the epidemiology of suicide in this group. Next, we review the interpersonal psychological theory of suicide (IPTS; Joiner, 2005) as a key theoretical explanatory model for the progression of suicide behavior. In order to identify how incremental steps toward suicide may develop in Bhutanese refugees, we propose an integration of the IPTS with the cultural model of suicide (Chu et al., 2010). We discuss the development of culturally-informed prevention efforts and share some practical lessons learned in conducting a pilot study of suicide risk detection in Bhutanese refugees resettled in Vermont.

Literature searches were conducted using PubMed Central databases and Google Scholar. Search stems included, but were not limited to, Bhutan, Refugee, Nepal, Immigrant, Culture, and paired with one or more of the following terms: Suici*, Ideation, Desire, Passive, Active. Because there is a sparse academic literature on the narrow topic of suicidality among refugees, broadly, and Bhutanese refugees, specifically, resources known to the authors were leveraged. Public health reports from the Centers for Disease Control and Prevention (CDC), aid and logistics organizations such as the International Organization for Migration (IOM), and U.S. states known to have large populations of Bhutanese refugees (e.g., Ohio). Journal articles citing any and all of these recourses were also included in our literature search.

History of Bhutanese Refugees

The history of discrimination of ethnically Nepalese (known as “Lhotshampa”) Bhutanese residents is substantial. Although a full account is beyond the scope of this review (the interested reader is referred to Hutt, 2003, 2005), a brief historical overview is presented in order to provide context for what follows. In the late 1800’s, Nepalese agricultural laborers migrated to southern Bhutan and settled in the region. These agricultural laborers were quickly joined by contracted laborers who were brought to the region through the early 1900’s (Hutt, 2005). These agrarian and contract workers and their decedents were allowed to speak Nepalese, practice their own religion (majority practiced Hinduism), and maintain their own cultural customs (Gulf Coast Jewish Family & Community Services: Refugee Services, 2015). In the late 1950’s, Bhutanese law granted full citizenship to Lhotshampa residents of Bhutan (Hutt, 2005). The governing bodies of Bhutan began enacting a series of discriminatory legislative actions starting in the late 1970s and early 1980s and escalating through the early 1990s when protests erupted following laws that restricted citizenship, linguistic, and cultural practices of ethnically Nepalese people living in Bhutan (Hutt, 2005). Violence, torture, and systematic oppression against the Lhotshampa became sanctioned by governmental bodies (Hutt, 2005). State-sponsored programs were enacted that restricted legal rights, access to education, and economic prosperity (Hutt, 2005). Ethnically Nepalese individuals from Bhutan were forced to flee to Nepal by way of India where they were not accepted as citizens (Hutt, 2005; Savada, Harris, & Library of Congress, 1993). In the mid-1990s, the Nepalese government attempted to negotiate with the Bhutanese government for the return of 80,000+ Bhutanese refugees living in Nepalese refugee camps; however, the Bhutanese government did not agree to readmit these refugees as citizens of Bhutan (Gulf Coast Jewish Family & Community Services: Refugee Services, 2015; Hutt 2005). By 2001, Bhutanese-sponsored terror and restrictions on basic rights led to 100,000 refugees living in seven refugee camps in southern Nepal. The Bhutanese refugee crisis escalated through 2008 when refugees began resettlement into third countries including the U.S., Canada, and Australia (Gulf Coast Jewish Family & Community Services: Refugee Services, 2015). Within the U.S. the largest populations of Bhutanese refugees reside in Pennsylvania, Texas, New York, Georgia, and Ohio, (Worldwide Refugee Admissions Processing System, 2018). As of January 2018, over 111,000 refugees resettled in eight different countries with a substantial majority resettling in the United (United Nations High Commissioner for Refugees, 2018). The Bhutanese refugee experience in the U.S. is unique and not easily generalized. However, a 2011 report (Shrestha, 2011) detailing the results of a 9-month anthropological study of the dynamics involved in Bhutanese refugee resettlement provides valuable insights as to the structural difficulties many Bhutanese refugees face upon resettlement. Among the issues raised by Shrestha (2011), is the mismatched expectations Local Refugee Resettlement Organizations (LRRO) and Bhutanese refugees have of one another. Chief among these mismatched expectations was the finding that many Bhutanese refugees in Shrestha’s study believed it was the responsibility of the LRRO to find them employment, while the LRRO believed their duty was limited to highlighting available resources for Bhutanese refugees to find their own employment (Shrestha, 2011). Shrestha (2011), highlights that inconsistent policies of the LRRO and variable discretionary aid provided by its staff often lead to mistrust of the LRRO among resettled refugees. Moreover, Shrestha (2011) noted that the asymmetric power dynamics of the LRRO can have a large impact on how LRRO staff view the populations they serve, contributing to problematic victim-savior dynamics. Finally, Shrestha (2011) detailed the challenges associated with integrating into a new cultural context, specifically, the othering and devaluation of the resettled population’s culture, customs, and needs. While these challenges exist anywhere refugees are resettled, in the state of Vermont, public opinion is largely in favor of refugees being resettled in the state (Bose & Grigi, 2017; the interested reader is directed to The Refugee Resettlement in Small Cities site [http://spatializingmigration.net] for great detail on how refugees are being received in the state of Vermont as well as ongoing projects examining how resettlement affects refugee populations).

Suicide-Related Terminology

To further set the stage for this review, we clarify our use of specific suicide-related terms. Within the past 10 years, a full hierarchical classification and nomenclature system for suicide-related behaviors has been developed and deployed (Silverman, Berman, Sanddal, O’Carroll, & Joiner, 2007a, 2007b). In this review, we adopt the consensus terms outlined in Silverman and colleagues (Silverman et al., 2007b) unless otherwise noted. A suicide attempt is characterized by some/non-zero intent to die. A suicide attempt involves three possible outcomes: no injury, non-fatal injury, and death. Along with efforts to develop a standardized nomenclature and classification system for suicide-related behaviors, efforts to standardize attributes of self-injurious thoughts such as suicidal ideation have also been made (Silverman et al., 2007b). In this review, suicidal ideation refers to thoughts of ending one’s own life or otherwise engaging in suicidal behaviors (Nock, Borges, Bromet, Cha, et al., 2008). Desire for death or suicidal desire – which is not included in Silverman et al.’s (Silverman et al., 2007b) classification system – is defined here as the longing for death by suicide (Van Orden, Witte, Gordon, Bender, & Joiner, 2008). Suicide plan is defined here as the specific premeditated methodology and design of an individual’s suicide.

Suicide among Bhutanese Refugees Resettled in the U.S.

Relative to the general U.S. population, Bhutanese refugees resettled in the United States are 2 times more likely to die by suicide (Ao et al., 2012). This alarming rate, based on data from the CDC (Ao et al., 2012) study, is likely an underestimate. Although the age-adjusted suicide rate of 24.4 per 100,000 people (Ao et al., 2012) is in line with the estimated suicide rates among forced migrants and refugees, broadly, recent reviews (Schininà, Sharma, Gorbacheva, & Mishra, 2011; Vijayakumar & Jotheeswaran, 2010) have highlighted highly variable suicide death rate estimates (ranging from 3.4% to 34%) across the literature and have suggested our ability to estimate suicide related behavior across refugees from different communities and regions is currently poor. A recent meta-analysis has shown that suicide rates of immigrants, broadly, are typically consistent with country of origin, not with host country (Voracek & Loibl, 2008). Unfortunately, in the case of Bhutanese refugees, neither Bhutan nor Nepal constitutes a true country of origin.

Examining the suicide rate in both Bhutan and the Nepalese refugee camps is necessary in order to fully contextualize the current suicide rate among Bhutanese refugees resettled in the U.S. The World Health Organization (WHO) estimated the age-adjusted suicide rate in Bhutan to be 13.9 per 100,000 people (World Health Organization, 2017). While in the Nepalese refugee camps, the estimate was 20.76 per 100,000, far greater than the rate in the surrounding Nepal region where the refugee camps were located (Schininà et al., 2011). These data show that Bhutanese refugees are at a significantly increased risk of suicide that remains unexplained by the suicide rate in their country of origin.

A recent study of Bhutanese refugees in Ohio (n=200) found that 21% of participants reported having a family member who died by suicide; 4X more than the 5% who endorsed the same question in the CDC’s study (Adhikari et al., 2015). Between 2009 and 2012 (the latest year in which aggregate data are available), there were 16 confirmed deaths by suicide among Bhutanese refugees living in the U.S.; seven of the confirmed suicides also had family members or close friends who died by suicide, suggesting a contagion-type effect (Ao et al., 2012). Moreover, all suicides occurred well within the first year of arrival in the U.S. (Ao et al., 2012). Current estimates of suicidal ideation among Bhutanese refugees are dramatically lower than expected given the high rate of suicide within this population (Adhikari et al., 2015; Ao et al., 2012). With an estimated 7,500 refugees remaining in Nepalese refugee camps (United Nations High Commissioner for Refugees, 2018), there is a crucial need for identifying culturally-sensitive, accurate, and early risk factors for suicidal behavior to aid in detecting individuals within the Bhutanese refugee population at risk for suicide.

The need to address the Bhutanese refugee suicide epidemic is recognized by agencies and researchers across the globe. Extensive work to gather quantitative and qualitative data on suicide risk factors among Bhutanese refugees living in the Nepalese refugee camps was conducted by IOM researchers Schininà and colleagues (2011). As part of their work, the group conducted psychological autopsies on a randomly selected group of nine fatal suicide attempters who made their fatal attempts in the Nepalese refugee camps. The IOM group also conducted clinical interviews with five individuals who attempted suicide in Nepalese refugee camps. Schininà et al. (2011) found that the most commonly reported risk factors of suicide attempts were untreated mental illness (especially major depressive disorder), alcohol abuse (not necessarily reaching clinical levels), and a history of suicide attempt(s; fatal or non-fatal) within their family. Within the camps, researchers found that virtually all fatal suicide acts were made by hanging using either rope or shawls, and that attempts were typically impulsive in nature. Another major risk factor among those interviewed by the IOM was a change in the role of family responsibilities (i.e., former providers no longer serving as providers). Additionally, feelings of shame and poor distress tolerance were cited as significant risk factors for suicide attempts among those interviewed.

A 2012 CDC study (Ao et al., 2012) replicated and extended many of the findings outlined in Schininà et al., (2011). The CDC also employed psychological autopsies in order to assess the circumstances surrounding suicides of Bhutanese refugees resettled in the U.S. Ao and colleagues (Ao et al., 2012) conducted psychological autopsies on 14 of 16 confirmed Bhutanese refugee suicides that occurred between 2009 and 2012. The demographic makeup of the sample suggests that Bhutanese males are at higher risk of fatal suicide attempts than females. Additionally, marital status and family provider status emerged as critical risk factors, with the majority of individuals who died by suicide married (74%) and not primary providers but traditionally expected to be in the role of primary provider (71%), replicating results from Schininà et al. (2011) . Religion and time in the U.S. also appeared to be significant risk factors; 79% of individuals with confirmed suicide deaths identified as Hindu, and the median time between arrival and death was 7.4 months for males and 1.1 month for females. This finding may mirror the finding in the broader U.S. population that the greatest risk of suicide occurs within the first 12 months following the onset of suicidal ideation (Kessler, Borges, & Walters, 1999; Nock, Borges, Bromet, Alonso, et al., 2008). As in the Schininà et al. (2011) study, nearly all suicides occurred by hanging (92%). However, perhaps most concerning for suicide prevention efforts, was the finding that 57% of respondents (i.e. family or friends) believed the suicide was impulsive and 46% did not recall signs that suicide was imminent. This suggests that prevention efforts relying on the presence of suicide plans and ideation prior to an attempt are likely to miss at-risk individuals.

As suicide fatalities can be reduced through proper identification of suicidal behavior and subsequent intervention (Mann et al., 2005), there is a critical need to identify at-risk individuals. Given that suicide risk in the U.S. is greatest within the first 12 months following the onset of suicidal ideation, prevention work typically aims to identify clinically useful risk factors that follow the onset of suicidal ideation, but precede suicide attempts (Kessler et al., 1999; Nock, Borges, Bromet, Alonso, et al., 2008). However, Bhutanese refugees may be “hidden ideators” (Morrison & Downey, 2000), evidenced by their unusually low rates of suicidal ideation (Ao et al., 2012; Adhikari et al., 2015) compared to the general U.S. population (13.5%) rate of suicidal ideation (Kessler et al., 1999; Nock, Borges, Bromet, Alonso, et al., 2008). Moreover, Bhutanese refugees living in the U.S. have a significantly elevated rate of suicide (Ao et al., 2012) when compared to that of the general U.S. population (National Center for Injury Prevention and Control: Division of Violence Prevention, 2015). These data highlight that risk assessments predicated on the presence of suicidal ideation would result in under-identification of those Bhutanese refugees at risk for suicide. Furthermore, suicidal ideation is stigmatized within the Bhutanese culture; therefore, relying on traditional risk factors (e.g., suicidal ideation) as the sole or primary indicator of suicide risk may not be responsive (Chu et al., 2010). Suicide screening tools in Bhutanese refugees require the inclusion of novel, more proximal, risk factors. Improved assessment and identification of Bhutanese refugees at-risk for suicide represents a significant empirical gap. A conceptual model of how incremental steps toward suicide develop in Bhutanese refugees is needed to develop culturally-informed prevention efforts.

Joiner’s Interpersonal-Psychological Theory of Suicidal Behavior

Until 2005, when the IPTS (Joiner, 2005) was published, models of suicide largely did not reconcile the high rates of suicidal ideation in the general population with the relative rarity of suicide attempts and deaths (Kessler, Berglund, Borges, Nock, & Wang, 2005; Kessler et al., 1999). The IPTS, however, posited that suicide-related behavior developed from the confluence of three factors: elevated social alienation (i.e., the perception that a person’s existence poses a burden to others), high levels of perceived burdensomeness (i.e., social/personal alienation from others), and increased acquired capacity for suicide (i.e., a decreased fear of death and increased pain tolerance gained through first-hand or second-hand experiences of painful and/or frightening stimuli). Inclusion of the latter factor was a novel feature in conceptual models of suicide at the time. According to the model, perceived burdensomeness and social alienation contribute to the concrete or abstract desire for death (i.e., desire to be dead or suicidal ideation), and planning.

Prior to Joiner’s IPTS, models failed to explain why most individuals who experienced suicidal ideation did not go on to attempt suicide (Klonsky & May, 2014; Nock, Borges, Bromet, Cha, et al., 2008). In the intervening years since Joiner’s model was first published, however, empirically-based models situated within the ideation-to-action framework, or the progression from suicidal ideation to suicide-related behaviors, have emerged (Klonsky & May, 2015; O’Connor, 2011) and continue to gain traction in the field (Burke & Alloy, 2016; May & Klonsky, 2016). These ideation-to-action-based models may help us contextualize the current suicide epidemic in the Bhutanese refugee community.

Since 2005, the IPTS model has gained widespread support in community samples (Christensen, Batterham, Soubelet, & Mackinnon, 2013; Van Orden et al., 2008), clinical samples (Joiner et al., 2009; Miller, Esposito-Smythers, & Leichtweis, 2015), and military samples (Bryan, Morrow, Anestis, & Joiner, 2010). A recent review (Hill & Pettit, 2014) of 27 empirical studies of the IPTS found that the core tenants of the model are strongly supported, with the relationship between perceived burdensomeness and both suicidal ideation and attempts consistently replicated in the literature. As the IPTS posits, the relationship between thwarted belongingness and suicidal ideation also has strong empirical support (Hill & Pettit, 2014). The review also found the dynamic nature of perceived burdensomeness to moderate the relationship between thwarted belongingness and acquired capability of suicide, which suggests that decreasing perceived burdensomeness may have significant and wide-reaching effects on suicide-related behaviors.

However, another recent systematic review suggests that the evidence for the IPTS is less clear than initially thought (Ma, Batterham, Calear, & Han, 2016). Ma and colleagues (2016) noted that perceived burdensomeness is unequivocally associated with suicidal ideation, with 82.6% of studies finding a significant association between the two constructs; however, the association between thwarted belongingness and suicidal ideation is less robust. A significant association between the latter two constructs was found in only 40% of studies, with thwarted belongingness contributing smaller amounts of variance in comparison to perceived burdensomeness (Ma et al., 2016). Although refinements to components of the IPTS may be necessary, the overall structure of the model is still supported, with a two-way interaction between thwarted belongingness and perceived burdensomeness predicting suicidal ideation in 66.6% of studies (Ma et al., 2016). While studies that tested the complete model are scant and likely underpowered, 42.8% of studies that tested the 3-way interaction of thwarted belongingness, perceived burdensomeness, and capacity for suicide found a significant association with suicide attempts. Ma and colleagues note that these mixed results may be the result of several factors including the IPTS failing to fully capture the true nature of the relationship among the model’s components, inadequate ability of certain measures (i.e., Interpersonal Needs Questionnaire) to capture thwarted belongingness, the questionable generalizability of IPTS studies to large heterogeneous samples, and other unaccounted-for variables that may better explain or moderate the relationships between the identified constructs (Ma et al., 2016).

Despite some mixed support for the IPTS, several of the constructs comprising the IPTS may apply well to the problem of suicide among Bhutanese refugees. One previous study of the IPTS among Bhutanese refugees reviewed ways in which thwarted belongingness and perceived burdensomeness may be contextualized by the experiences of Bhutanese refugees, specifically highlighting that physical distance from family members, language and cultural differences between resettled refugees and the host communities, and new status as a minority outgroup after 1.5–2 decades in refugee camps with fellow group members may all contribute to the construct of thwarted belongingness (Ellis et al., 2015). Thwarted belongingness may also be exacerbated by restricted availability of cultural and religious centers (Ao et al., 2012). Perceived burdensomeness also has roots in the Bhutanese refugee experience. Ellis and colleagues (2015) explain that Bhutanese refugees may experience downward social mobility in the new host country, and they may be unable to fulfil culturally-accepted familial roles due to structural barriers (e.g., not speaking the host-country’s language and difficulties obtaining employment). Furthermore, they highlight that the collectivist culture of Bhutanese refugees may magnify the negative psychological impact of being reliant on others (reviewed in Ellis et al., 2015). Given the high level (22.4%) of psychiatric disability among Bhutanese refugees (Thapa, Van Ommeren, Sharma, de Jong, & Hauff, 2003), it is also possible that many Bhutanese refugees require higher levels of care from state agencies, family members and friends. Psychiatric disability status may, subsequently, be associated with perceived burdensomeness. Another significant strength of the IPTS is that it defines suicidal desire and suicidal ideation broadly, which may allow for much needed flexibility when defining the construct for a population in which there is a high level of stigma around expressing suicidal ideation. Moreover, given that Bhutanese refugees tend to display unusually low rates of suicidal ideation (Adhikari et al., 2015; Ao et al., 2012), the broad definition of suicidal ideation in the IPTS may make it an ideal candidate model with which to examine suicide among Bhutanese refugees.

Ellis and colleagues (2015) examined elements of the IPTS, among Bhutanese refugees. To our knowledge, this is the only published study to date attempting to demonstrate the utility of a theoretical model in explaining suicide among Bhutanese refugees. Using data collected as part of a CDC study (Ao et al., 2012), Ellis and colleagues (2015) examined participants’ physical health, psychiatric diagnostic status, traumatic experiences, post-migration challenges, social support, and IPTS-consistent measures of perceived burdensomeness and thwarted belongingness. Among other risk factors that replicate previous research, Ellis and colleagues (2015) showed an association between IPTS-consistent risk factors and suicidal ideation. Specifically, they found that Bhutanese refugees who endorsed suicidal ideation presented with higher perceived burdensomeness and lower interpersonal belonging than Bhutanese refugees who did not express suicidal ideation. However, one limitation of Ellis and colleagues’ work was that suicidal ideation was the only indicator of suicide risk examined. Researchers asked participants whether or not they had ever seriously thought about suicide and only asked probing questions upon an affirmative response. The effect of this procedure may be a failure to capture the full spectrum of suicidal ideation and subsequent risk of suicidal behavior in the sample, which reported an unusually low rate (3%) of suicidal ideation, yet the general population of Bhutanese refugees experience an extremely high rate of suicide (Ao et al., 2012).

Chu, Goldblum, Floyd, and Bongar’s Cultural Model of Suicide

In response to myriad cultural variation in the presentation of suicide risk, especially in minority cultural groups, Chu et al. (Chu et al., 2010) developed a theoretical model that is culturally versatile and comprehensive. The cultural model of suicide is an empirically-based theoretical framework that aims to bring a flexible definition of several key constructs that, together, are hypothesized to account for a significant majority of culturally-specific suicide risk. Constructed by inductive methods after thorough literature reviews, the cultural model of suicide consists of four key factors: (1) cultural sanctions, or the extent to which a behavior (inclusive of suicide) is considered culturally acceptable or unacceptable, (2) idioms of distress, or the patterns and behaviors through which distress is expressed or communicated, (3) minority stress which consists of experienced stress that results from any aspect of minority status, and (4) social discord which consists of any aspect of interpersonal conflict or social and cultural isolation (Chu et al., 2010). The cultural model of suicide posits that these four factors, in combination and in relation to one another, contribute to the development of suicidal behavior. Three organizing principles outline the relationship among each of the four previously defined factors: (a) cultural factors dictate the expression of suicidal ideation, intent, plans, acts, and methods; (b) these cultural factors dictate which life stressors and moderating risk and protective factors ultimately lead to suicide-related thoughts and behaviors, (c) cultural factors dictate the meaning of stressful life events suicidal thoughts and/or behaviors. According to the cultural model of suicide, stressful events occur and are interpreted through a cultural lens, if the event and/or response to the stressful event are culturally accepted, then the event is generally tolerated and coped with. However, if the event and/or response to the event is culturally unacceptable or exceeds coping capacity, culturally-bound idioms of distress develop (ideation, plans, intent). If expressions of suicide are culturally unacceptable this can contribute to increased tolerance of negative affect, but if suicide is culturally acceptable, this may lower the threshold of tolerance and increase risk of suicidal behaviors. Once the threshold of tolerance is exceeded, the cultural model of suicide posits that one is likely to engage in culturally mediated suicide-related behavior.

In the short time since the development of the cultural model of suicide, the four key factors and organizing framework have gained support and have been incorporated into clinically useful, and psychometrically valid, measures: The Cultural Assessment of Risk for Suicide (CARS; Chu et al., 2013) and the CARS-S (Chu, Hoeflein, Goldblum, Espelage, et al., 2017), which is a shortened screener version of the same measure, assess elements of the four theoretical constructs of the cultural model of suicide (i.e., cultural sanctions, idioms of distress, minority stress, and social discord). Though the cultural model of suicide has not been subjected to a large-scale test of the theoretical framework, it has been shown to be clinically useful in a case study of an individual with multiple overlapping and intersecting cultural identities (Chu, Hoeflein, Goldblum, Bongar, et al., 2017). The cultural model of suicide model is still emerging and has limited empirical support because it is new and dissemination throughout the field of suicidology takes time. However, there is promising evidence supporting each of the model’s key assertions, namely that cultural factors influence the experience and interpretation of life stressors, the expression of suicidal thoughts and behaviors, and the meaning created from such events (Chu et al., 2010). While the cultural model of suicide has emerging empirical support, the underlying constructs are well supported throughout the literature and serve as a promising model of suicide.

No specific applications of the cultural model of suicide have, as of yet, been developed for the resettled Bhutanese refugee community; however, many of the constructs that Chu and colleagues (2010) developed translate well to this population. Cultural sanctions, for example, are clearly present within the Bhutanese refugee community. Suicide is highly stigmatized among Bhutanese refugees (Adhikari et al., 2015; Ao et al., 2012). Moreover, while living in the Nepalese refugee camps suicide was illegal, and, according to Hindu tradition (the religious affiliation of a majority of Bhutanese refugees), was considered to be a sin, though not an offense that would prevent proper burial while according to Buddhist tradition suicide is considered a highly discouraged action, but is not explicitly outlawed (Schininà et al., 2011). Of 16 families surveyed while living in the refugee camps, 10 indicated that suicide was an understandable course of action, but not an acceptable one (Schininà et al., 2011).

Idioms of distress have also been documented as uniquely culturally defined in the Nepali and Bhutanese refugee communities (Kohrt & Hruschka, 2010; Sharma & Ommeren, 1998). An IOM report (Schininà et al., 2011) highlighted that of individuals from 16 families surveyed in the Nepalese refugee camps, 23.2% who were not related to someone who engaged in a fatal or nonfatal suicide attempt said that suicide was related to dimaag (brain–mind/serious psychological disturbances), and 55% of individuals who were related to someone who engaged in a fatal or nonfatal suicide attempt said that suicide was related to dimaag. In contrast, 30.4% of individuals who were not related to someone who engaged in a fatal or non-fatal suicide attempt said that suicide was related to man (Heart-Mind/emotional disturbances) while only 25% of individuals who were related to someone who engaged in a fatal or non-fatal suicide attempt used man to explain their family member’s suicide (Schininà et al., 2011). These idioms of distress are used to explain suicidal actions and behaviors differently depending on how relationally close someone was to a suicide attempter. Moreover, another idiom of distress may include the culturally-specific means by which individuals choose to die by suicide. As of 2012, 81.25% of fatal suicide attempts by resettled Bhutanese refugees occurred by hanging (Ao et al., 2012).

Social discord is another construct of the cultural model of suicide that maps on to the problem of suicide among Bhutanese refugees. In an Ohio department of health study (Adhikari et al., 2015), 5% of resettled Bhutanese refugee men and 4% of Bhutanese refugee women experienced domestic violence, however due to cultural factors that place the importance of the family unit above individual safety, this is likely a significant underestimate (Wang, 2017). Moreover, increased family conflict was significantly associated with past suicidal ideation (Ao et al., 2016). Another aspect of social discord involved perceived social support and connections. Among Bhutanese refugees, lower perceived social support was significantly predictive of suicidal ideation (Ellis et al., 2015). IPTS constructs such as thwarted belongingness also fall into the sub-category of social discord and as stated previously, thwarted belongingness was significantly associated with suicidal ideation (Ellis et al., 2015). With regard to minority stress, Bhutanese refugees resettled in the US experience elevated levels of postmigration living difficulties and acculturative stress (Adhikari et al., 2015; Ao et al., 2012), which are associated with past (Ao et al., 2016) suicide ideation.

An Integration of the Cultural Model of Suicide and the IPTS

While the four key factors of the cultural model of suicide have a strong empirical basis that translate well for the specific use-case of modeling risk of suicide within the Bhutanese refugee community, the causal relationships of the cultural model of suicide have not yet been tested in a rigorous large-scale study. A large-scale validation study of the assumptions of the cultural model of suicide would be an exciting and important next step. The IPTS is a well-validated explanatory framework. The constructs comprising the IPTS are flexible enough to be tailored to a particular cultural context. Moreover, thwarted belongingness and perceived burdensomeness already have strong empirical support within the Bhutanese refugee community (Ellis et al., 2015). Given that in many minority cultures it is common for suicidal ideation to be suppressed and present as “hidden ideators” (reviewed in Chu et al., 2010), choosing a framework that focuses on suicidal ideation may seem to be contraindicated, however, by integrating socio-cultural adaptations into the ideation-to-action framework, and by defining ideation broadly, the ideation-to-action framework may provide useful and relevant information about the progression to, and incremental risk of, suicidal behavior. Subsequently, a culturally-responsive adaptation of the IPTS that includes moderators based on the cultural model of suicide will provide a framework for examining incremental suicide risk and the progression from distal and proximal risk factors to suicidal behavior.

We propose a culturally-responsive conceptual model for the development of suicide-related behavior consistent with the ideation-to-action framework. Our model flexibly defines constructs such as thwarted belongingness to incorporate select indicators of social discord such (i.e., social support) based on the cultural model of suicide in addition to traditional thwarted belongingness indicators. The high prevalence of psychiatric disability experienced by Bhutanese refugees (Thapa et al., 2003) among other factors may be associated with higher levels of perceived burdensomeness, subsequently, our model also includes traditional IPTS indicators of perceived burdensomeness. Finally, given that Bhutanese refugees exhibit characteristics of “hidden ideators” (reviewed in Chu et al., 2010), namely low levels of explicit thoughts about suicide, the IPTS with its broad definition of suicidal ideation/suicidal desire is a particularly compelling foundational model as suicidal desire may account for the low levels of reported suicidal ideation that has, in previous studies, been narrowly defined. Relative to other theoretical models, the IPTS has strong preliminary empirical support within the context of the Bhutanese refugee community (Ellis et al., 2015).

Our proposed model (Figure 1) includes the abstract desire to be dead (as originally proposed by Joiner) in an effort to capture the risk of suicidal behavior better than suicidal ideation in the Bhutanese refugee community. We propose that, within the Bhutanese refugee population, which appears to experience low levels of narrowly defined suicidal ideation, the interaction of perceived burdensomeness and thwarted belongingness also predicts higher desire for death in addition to suicidal ideation. We further propose that desire for death captures additional risk of suicide-related behavior above and beyond suicidal ideation alone in this population. Based on the IPTS, suicidal desire is insufficient for lethal or near lethal suicide events. A third component, the acquired capability of suicide is necessary, according to the theory, in order for a lethal or near lethal suicide attempt to occur (Joiner, 2005; Van Orden et al., 2008). Subsequently, suicidal ideation and desire for death are hypothesized to be related to suicidal behaviors, but their combination is not hypothesized to contribute additional risk of suicidal behavior. According to the IPTS, the moderating factor would be an independent construct: acquired capability of suicide. The proposed conceptual model provides a beginning framework for a culturally-responsive model of the development of suicide among Bhutanese refugees.

Figure 1:

Conceptual adaptation of the Interpersonal Psychological Theory of Suicide and the Cultural Model of Suicide with additional moderators in Bhutanese refugees

Note: Moderators displayed in grey rectangles. Dotted Lines represent inclusion of one or more elements of the Cultural Model of Suicide (Chu et al., 2010)

Our model differs slightly from Joiner’s IPTS. Joiner’s IPTS posits that suicidal ideation is the operationalization of suicidal desire; however, a low level of reported suicidal ideation preceding suicide-related behavior in Bhutanese refugees requires us to broaden the definition of ideation in the context of the ideation-to-action framework. Our application of the IPTS includes desire for death, the longing for death, a construct that has been shown to be modestly related to but distinct from suicidal ideation (Dadfar, Lester, & Vahid, 2016; Dadfar et al., 2016; Lester, 2013), and may – in the case of “hidden ideators” like Bhutanese refugees – better capture the broad construct of suicidal desire and suicidal ideation that Joiner first proposed. By distinguishing between these two related constructs, our conceptual model may aid in establishing proximal risk factors of suicidal behavior and identifying the earliest and most accurate warning signs for suicide risk in Bhutanese refugees to date. Translational science based in the ideation-to-action framework can yield clinically useful risk factors for suicidal behavior. Our proposed application of the IPTS and cultural model of suicide will lay the groundwork for future longitudinal tests of the proposed risk factors for suicidal behavior and will allow for the development of interventions designed specifically to reduce suicidal behavior among Bhutanese refugees.

Potential Moderating Variables

In reviewing the literature on suicide among Bhutanese refugees, it is clear that numerous additional factors outside of thwarted belongingness, perceived burdensomeness, and acquired capability of suicide play a significant role in the progression from ideation to action. Specific moderating factors can impact the degree to which an individual is at risk for developing suicidal ideation, desire, or behavior. Trauma exposure, post-migration living difficulties and minority stress, idioms of distress, cultural sanctions, are several such potential moderating variables. Additionally, several protective factors might partially mitigate the effects of perceived burdensomeness, thwarted belongingness, desire for death, and suicidal ideation.

Trauma exposure.

Trauma and torture are considered to be risk factors for general psychopathology in Bhutanese refugees and may represent risk factors for suicide as well (Ferrada-Noli, Asberg, & Ormstad, 1998; Ferrada-Noli, Asberg, Ormstad, Lundin, & Sundbom, 1998; Hovey, 2000). One of the most pressing aspects for any refugee crisis spurred by systematic oppression is the likelihood that refugees have experienced torture, poor living conditions, and general trauma exposure before and after fleeing their home countries as well as while living in refugee camps. Trauma exposure is a significant risk factor for eventual psychopathology and other medical and health related problems, in particular, disorders such as posttraumatic stress disorder (PTSD) and depression. In a sample of Bhutanese refugees resettled in the U.S., 21% reported symptoms consistent with major depressive disorder (MDD), 4% reported symptoms consistent with PTSD, 18% reported generalized anxiety symptoms, and 45% reported experiencing symptoms of more than one disorder (Vonnahme, Lankau, Ao, Shetty, & Cardozo, 2015).

Specific types of profound trauma, such as torture are particularly prevalent among Bhutanese refugees resettled in the U.S. and, depending on the sample, estimates are wide ranging, from 15% (Ao et al., 2012) to 95% (Adhikari et al., 2015). Torture is defined by the United Nations Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment as the intentional infliction of mental or physical pain in order to intimidate, coerce, elicit information, or simply as a result of discrimination. Torture is inflicted or endorsed by states, officials, or their designees and excludes inherent or incidental pain resulting from sanctioned lawful actions (United Nations, 1984).

Torture status among Bhutanese refugees was studied in one Nepali refugee camp (Shrestha et al., 1998; Van Ommeren et al., 2001). In a comparison of 526 Bhutanese refugee torture survivors and 526 non-tortured Bhutanese refugees on a number of psychiatric sequelae, torture status significantly predicted current psychiatric disorders including PTSD, MDD, and generalized anxiety (Shrestha et al., 1998). A follow-up study two years later (Van Ommeren et al., 2001) with the same participants (torture survivors: n=418; non-tortured: n=392) using updated diagnostic criteria replicated and extended previous findings, showing that torture status continued to significantly predict psychiatric disorders. Specifically, torture status predicted lifetime rates of posttraumatic stress, pain, generalized anxiety, dissociative, and affective disorders. Torture status also predicted 12-month prevalence of posttraumatic stress, pain, and dissociative disorders but – notably – not mood or anxiety disorders (Van Ommeren et al., 2001). Taken together, these data suggest that torture status at least partially influences the development of severe psychiatric disorders which can lead to significant impairment.

Relative to non-tortured Bhutanese refugees, tortured Bhutanese refugees experience more psychiatric disorders and often have more co-morbid psychiatric disorders (Van Ommeren et al., 2001). Ellis and colleagues (2015) found that psychopathologies such as anxiety and depression, but not PTSD, were significantly associated with suicidal ideation among Bhutanese refugees. Perhaps related to the lack of a significant relationship between PTSD and suicidal ideation in their study is the finding that Bhutanese refugees with and without suicidal ideation had no significant differences in the number of traumas they experienced (Ellis et al., 2015). Ellis and colleagues (2015) concluded that trauma exposure alone does not explain the high rates of suicide among Bhutanese refugees. They hypothesized that trauma exposure may instead contribute to an increased risk of an IPTS construct called acquired capability of suicide which would affect the progression from suicidal ideation to suicidal action (Ellis et al., 2015). Based on the evidence presented, we echo Ellis and colleagues (2015) hypothesis and stipulate that trauma and torture may significantly moderate the transition from suicidal ideation and desire to be dead to suicidal behavior.

Post-migration living difficulties and minority stress.

Throughout the literature on refugee resettlement, one of the key elements that emerges consistently is the intense struggle new immigrants face with respect to post-migration living difficulties. Post-migration living difficulties are barriers one faces after resettlement or immigration of any kind. The post-migration living difficulty construct is largely consistent with the minority stress construct from the cultural model of suicide. In the cultural model of suicide, sexual minority stress, acculturative stress, and non-specific minority stress are measured factors within the subsuming minority stress construct (Chu et al., 2013). As such, post-migration difficulties and minority stress contribute to the same theoretical construct and, for the purposes of this study, will be used interchangeably. Examples of post-migration stressors include language and employment-related problems, cultural differences between the immigrant and the host culture, discrimination, and pragmatic issues such as proximity to transportation, among others. Prior studies have shown that post-migration living difficulties are both highly prevalent among samples of Bhutanese refugees resettled in the U.S. and highly related to psychopathology and suicide among the resettled Bhutanese refugee population (Adhikari et al., 2015; Ao et al., 2012). Highlighting the significant impact minority stress can have on one’s quality of life, one qualitative study of older Bhutanese refugees resettled in the U.S. noted that key stressors included the challenges of navigating religious customs without established Hindu temples or priests and not speaking the language of the host culture, which translated into economic and employment-related challenges (Gautam, Mawn, & Beehler, 2017). Gautam et al. (2017) also found that loneliness following resettlement was a significant stressor in older Bhutanese refugees, many of whom transitioned from being surrounded by familiar faces to living in areas where they do not have established social connections.

Other studies of Bhutanese refugees have consistently shown minority stressors are associated with psychopathology and suicidal ideation. Specifically, worries about family members who remain in refugee camps, financial constraints, feelings of a loss of choice over the future, poor community structures, and the challenge of maintaining cultural customs were all significantly related to depressive symptoms in one study (Vonnahme et al., 2015). While similar stressors were significantly related to suicidal ideation, the specific post-migration stressors that were most strongly associated with suicidal ideation included increased family conflicts (an indicator that is also consistent with social discord according to the cultural model of suicide), employment opportunity challenges, inadequate access to mental health services, lack of choice over one’s future, minimal government assistance, and poor structures in the host culture to resolve domestic and familial disputes (Ao et al., 2016). The relationship between minority stress, psychopathology, and suicidal ideation suggest post-migration difficulties may play a moderating role in the development of suicidal ideation. Moreover, these same post-migration stressors were shown to be associated with elements of the IPTS including thwarted belongingness and perceived burdensomeness, though the relationships between individual stressors and IPTS constructs were moderated by gender (Ellis et al., 2015). Higher thwarted belongingness was associated with poor access to healthcare, being unable to find work, lack of choice over one’s future, separation from family, worries about family back home, and increased family conflict among women, but not men. While higher thwarted belongingness among men, but not women, was associated with an inability to pay living expenses. Higher perceived burdensomeness was associated with poor access to healthcare, lack of choice over one’s future, inability to pay living expenses, separation from family, and increased family conflict among women, but not men. While difficulty maintaining cultural and religious traditions was the only post-migration living difficulty associated with higher levels of perceived burdensomeness among men, but not women (Ellis et al., 2015). Ellis and colleagues (2015) hypothesized that many of the moderating effects of gender on post-migration stressors and their subsequent relationship to IPTS constructs stem from the existence of prescribed traditional gender roles of men and women. Ellis et al. (2015) highlighted that women in Bhutanese culture were primarily responsible for raising children and caring for the home and had limited ability to pursue education or other activities based outside the home or family-unit. Once resettled, families were geographically isolated from relatives and fellow Bhutanese refugee community members. Families were placed in a position where both men and women were required to earn income, which violated traditional gender norms (Ellis et al., 2015). Moreover, women were subjected to disproportionate levels of gender-based violence during migration and in the camps and experienced increased levels of gender-based trauma (reviewed in Ellis et al., 2015). Ellis and colleagues (2015) noted that Bhutanese men, traditionally, served as the primary financial provider for the family and served as the head of the household, however, in the U.S. post-migration stressors such as language barriers meant increased difficulties financially providing for one’s family. While these hypotheses are not exhaustive or explanatory, they provide an important dimensionality to the experience of minority stress and highlight that gender moderates minority stress and as such should be included in our model.

From the data presented, it is clear that minority stress moderates the level of experienced thwarted belongingness and perceived burdensomeness as well as the experience of suicidal ideation. We hypothesize that post-migration living stressors, moderated by gender, will impact the relationship between thwarted belongingness and suicidal ideation and desire for death, as well as the relationship between perceived burdensomeness and suicidal ideation and desire for death. Increased post-migration stressors will lead to increased levels of desire for death and suicidal ideation.

Social discord.

Drawing on the cultural model of suicide suggests that social discord, consisting of two sub-factors: family conflict and social support, play a significant role in the development of suicidal behaviors (Chu et al., 2010). In Chu and colleagues (2010) review of the literature, they note extensive evidence that suicide is related to family conflict, especially in Asian American and collectivist-oriented populations, this finding mirrors findings in the Bhutanese refugee suicide literature that highlight the significant relationship between family conflict and past suicidal ideation (Ao et al., 2016). An earlier study, in the same sample, of perceived burdensomeness and thwarted belongingness found that increased family conflict was predictive of significantly higher levels of perceived burdensomeness and thwarted belongingness in women, but not men (Ellis et al., 2015). Highlighting the important role gender may play in moderating the effects of family conflict. These gender differences moderating the relationship between family conflict and IPTS constructs may be the result of the traditional gender roles outlined earlier, specifically, that Bhutanese women, traditionally, cared for children and households and were more likely to be victims of gender-based violence. Subsequently, increased family conflict likely meant that this conflict was experienced more directly by Bhutanese refugee women than it was by men. We hypothesize that increased family conflict plays a significant role in moderating the level of both perceived burdensomeness and thwarted belongingness, moreover, we expect that these moderating effects are experienced to a greater extent by women than men.

Idioms of distress.

Expressions of suicide-related distress (e.g., suicidal ideation) are moderated by cultural factors according to the cultural model of suicide (Chu et al., 2010). The expression of suicidal thoughts, means chosen to die by suicide, and the likelihood that suicidal thoughts or desires are expressed, according to Chu et al. (2010) are hinged on the cultural norms, practices, and experiences. Chu and colleagues (2010) point out that “hidden ideation” is a prime example of the cultural variation in idioms of distress. They argue that for groups that are “hidden ideators,” distress is expressed in many other ways such as impulsivity, risk-taking behaviors, gambling, substance use, etc. Moreover, Chu and colleagues (2010) argue that cultural variation will dictate the methods and means chosen to die by suicide. Subsequently, these culturally specific idioms of distress may aid in explaining “hidden ideation” among Bhutanese refugees as well as the overwhelming majority of suicide deaths by hanging. Thus, including idioms of distress as a crucial moderator of suicide ideation and suicidal behavior will significantly increase the ability of our model to detect and contextualize non-traditional risk factors for suicide.

Cultural sanctions.

The acceptability of certain behaviors such as suicide and the subsequent unacceptability of responses to stress can, as reviewed by Chu et al. (2010), play a major role in determining the level of suicide risk. Specifically, Chu and colleagues (2010) note that the extent to which suicide is condoned or condemned in one’s culture can influence the likelihood that one makes a suicide attempt, subsequently affecting one’s risk. In the context of Bhutanese refugees, these cultural sanctions come in the form of the illegal status of suicide in the Nepalese refugee camps as well as the religious proscription of suicide in Hinduism and the strong condemnation of suicide in Buddhism (Schininà et al., 2011). These cultural sanctions, may subsequently serve as protective factors against suicide, and may mitigate the likelihood that one who experiences suicidal desire and/or suicidal ideation engages in suicidal actions. However, a majority of families surveyed in the Nepalese refugee camps (Schininà et al., 2011) noted that suicide was understandable even if it was not acceptable. Schinina et al. (2011) go on to note that in their small sample, older Bhutanese refugees appeared to take a stricter stance on the unacceptability of suicide than younger generations. Though speculative in nature, it is conceivable that the origins of this generational split may represent a difference in a number of factors including social support, hopelessness, closeness to those who have died by suicide, and/or more rigid ties to one’s religious strictures. Cultural sanctions clearly have a role in moderating the progression of suicidal behavior. Subsequently, it is crucial that we include the construct in our conceptual model.

Protective factors.

Protective factors such as certain religious orientations, social support, and attitudes towards a political conflict have been shown to moderate some of the effects of torture and trauma and to mitigate psychopathology more broadly (Vonnahme et al., 2015). Consistent with a cognitive model of depression, in a post-conflict region of Nepal, individuals who held more negative beliefs about the disruptive impact of the political conflict were more likely to develop depressive symptoms than those with less negative views of the conflict (Luitel et al., 2013). This suggests that cognitive biases (Beck, 1963, p. 196) likely moderate the impact of triggering events. This may extend to resettled refugees whereby individuals with relatively low cognitive vulnerability may be protective against perceived burdensomeness and thwarted belongingness.

Religious involvement, Buddhism specifically, was shown to be protective against anxiety and depression (Shrestha et al., 1998). In addition to potentially providing spiritual fulfilment, religious beliefs and ceremonies may provide social support and spiritual coping. Furthermore, a temple or house of worship may provide a community gathering place. These benefits of religious orientation may help mitigate the negative effects of psychopathology and development of suicidal desire or ideation. As it relates to suicide and our conceptual model, it is possible that religious orientation moderates the relationships between thwarted belongingness and suicidal ideation and desire for death as well as the relationship between perceived burdensomeness and suicidal ideation and desire for death. A recent opinion article (Chase, 2012) highlighted that connections to community organizations may also influence the extent to which individuals experience social alienation and/or its negative effects. The article specifically discusses the initiatives of community organizations in Burlington, Vermont as exemplifying a well-connected Bhutanese refugee community that prioritizes community organizations and gatherings in order to foster resilience (Chase, 2012). Based on empirical and anecdotal findings, supporting and cultivating protective factors that promote resilience through therapeutic interventions, may (at least partially) mitigate the effects of trauma, torture, and psychopathology and may have a significant role in moderating the progression from suicidal ideation to action.

Prevention and Recommendations

Prevention of suicide is the ultimate goal in the search for risk factors. Enabling providers, clinicians, and community members to be able to identify warning signs and take action to ensure safety is of primary importance. A comprehensive review by Mann and colleagues (2005) identified key elements of effective suicide prevention. Proper and frequent screening is among the best methods of identifying individuals who are at risk for suicide. Between 45% (Luoma, Martin, & Pearson, 2002) and 66% (Andersen, Andersen, Rosholm, & Gram, 2000) of individuals who go on to die by suicide had contact with a medical professional within the month before their deaths suggesting that better screening is critical in order to identify those at risk. We specifically recommend using measures that do not exclusively rely on the presence of suicidal ideation especially when working with minority populations as there is evidence to suggest that cultural idioms of distress for many minority populations do not include explicit suicide ideation (Chu et al., 2010). Using a measure such as the Wish to be Dead Scale (Lester, 2013) as well as the Cultural Assessment of Risk for Suicide (CARS; Chu et al., 2013) or screener version (CARS-S; Chu, Hoeflein, Goldblum, Espelage, et al., 2017) may provide additional information to clinicians about risk of suicide. Additional recommendations from Mann and colleagues (2005) include improved screening tools and increased frequency across all care-related points of contact for suicidal individuals. For example, the Substance Abuse and Mental Health Service Administration (SAMHSA) initiated the Screening, Brief Intervention, and Referral to Treatment (SBIRT) system in order to effectively identify, treat, and secure appropriate care for individuals who present as “at-risk” for a variety of mental health and substance-related problems. In Vermont, the SBIRT program relies on training medical staff, case managers, trusted cultural leaders, and other “gatekeepers.” Gatekeepers are leaders in the community that may have regular contact with individuals who may be at risk, and may include religious leaders, community elders, teachers, pharmacists, and institutional staff at all levels. In some resettled Bhutanese refugee communities, there is hesitation to seek medical care unless absolutely necessary as well as a reliance on traditional healers who play a substantial role in medical care (reviewed in Centers for Disease Control and Prevention, 2014). Establishing a public health model of suicide prevention that engages and trains all stakeholders and gatekeepers (e.g., traditional healers, elders, and religious leaders) to detect suicide risk will allow for earlier and more effective intervention. Within the Bhutanese refugee community, religious leaders may be especially well positioned to serve as gatekeepers given that cultural sanctions and religious orientation may have a strong moderating role on suicidal ideation and behavior (Chu et al., 2010; Ellis et al., 2015). Mann and colleagues (2005) also noted the importance of developing good evidence-based psychiatric and psychological care. This care is crucial in order to effectively manage and treat suicidality as well as underlying factors contributing to suicidality. Clear policies and procedures in healthcare settings for how and when to engage patients in effective follow-up care (e.g., phone calls or text messages upon discharge from hospitals and psychiatric units), can also significantly improve outcomes and reduce deaths (Chung et al., 2017). Means restriction techniques such as fences, nets on bridges, gun locks, and safe drug and prescription disposal receptacles can also substantially reduce incidences of suicide-related deaths (Yip et al., 2012). Given that most Bhutanese refugees who die by suicide die by hanging, establishing a public health campaign to empower stakeholders to identify culturally-appropriate warning signs (idioms of distress) of a mental health crisis as well as de-stigmatize the temporary restriction of access to, or removal of, culturally-specific means of hanging from one’s home may be a worthwhile endeavor. Additionally, responsible media reporting and public education about how to avoid social media posts that sensationalize suicide deaths can significantly mitigate the risk of suicide clusters and contagion (Bohanna & Wang, 2012). Finally, developing frequent and engaging community programing that fosters connection among community members, facilitates healthy community building within the Bhutanese refugee community, and also integrates non-Bhutanese refugee community members may mitigate the experience of isolation, thwarted belongingness, perceived burdensomeness, and social discord. Providing economic opportunities such as grants, credit, and loans for refugees to develop their own self-sustaining businesses, increasing partnerships with local companies to hire refugee workers, integrating English Language Learner programs into job settings, and providing easy access to transportation may be important host-community goals that mitigate perceived burdensomeness of refugee community members.

Mann et al. (2005) argued for a collectivist, collaborative, and public health approach to suicide prevention in which all members of a regional or national community are stakeholders that have a responsibility to promote the health and safety of fellow stakeholders. This type of approach places the onus of prevention on all stakeholders. This approach contrasts a medical-oriented model in which individuals are treated only if and when symptoms are displayed (U.S. Department of Health and Human Services, 2010). Mann and colleagues (2005) argued that a collectivist and collaborative model that engages community stakeholders in prevention efforts will be an effective tool that will facilitate detection and intervention early. This collaborative approach to suicide prevention has been applied in American Indian and Alaska Native communities (e.g., U.S. Department of Health and Human Services, 2010) and is embodied by the Zero Suicide initiative (Hogan & Grumet, 2016). The Zero Suicide initiative is a systems-based approach that has shown early promise for being an effective tool for reducing deaths from suicide and increasing suicide prevention in healthcare-oriented organizations. The approach integrates seven key elements including: (1) a commitment from leadership to reduce deaths from suicide; (2) proper training for staff so that they are adequately prepared to assess and care for suicidal individuals; (3) regular use of evidence-based assessment tools and screenings; (4) a structured and unambiguous care protocol for caring for suicidal individuals; (5) evidence-based treatment for suicidality specifically; (6) increased support during transitions of care for suicidal individuals; and (7) outcome measurement and data review to identify problems and inefficiencies.

Although these principles are important for organizations both large and small to take into account, there are elements that may allow these initiatives to be successful across cultures (Alegría, Alvarez, & Falgas-Bague, 2017). Among these cross-cultural initiatives are increasing language accessibility resources such as translation and interpretation services, and multiple pathways into services such as self-referrals or provider referrals. Additionally, it is essential to proactively build partnerships with community organizations such as religious temples, community centers, case management services, and local lay-lead groups in order to conduct jointly sponsored outreach events to promote services, healthy living, and mental health awareness and resources. Working collaboratively with community leaders in an egalitarian model of mutual respect will allow for effective dissemination of important information, screening, and serves to be an excellent outreach tool. Another tremendously useful tool for cross-cultural effectiveness is developing an interdisciplinary, cross-cultural, and inter-organizational care-net where service organizations, community leaders, local politicians, and government agency’s gather regularly to disseminate updates, introduce communities to new initiatives, coordinate community resources, and maintain active and up-to-date knowledge about major community events and emerging community health crises. Through care, coordination, and constant improvement of the science that governs policies and procedures, improved suicide prevention is possible.

Pragmatic Lessons Learned

Suicide is a significant public health challenge across the globe, with prevention efforts in full force and some governments actively trying to reduce the number of suicides in their countries. While Nepali-speaking Bhutanese refugees do not generally return to Bhutan, the Bhutanese government has developed a three-year action plan to reduce the number of suicides occurring in the country. Their action plan consists of six primary objectives: (1) Encourage leadership in the community on issues related to suicide prevention (e.g., clergy leadership, local government awareness campaigns, school awareness/prevention initiatives, responsible media reporting, etc.); (2) Create national government divisions and institutions with the specific focus of preventing suicide deaths; (3) Create better access to prevention services and increase crisis support; (4) Establish standardized trainings to ensure medical professionals and other gatekeepers meet practice guidelines; (5) Strengthen community resilience by increasing suicide awareness outreach in local communities and increasing availability of counseling support services; (6) Improve data collection and analysis standards around suicide-related issues.

The response of the Bhutanese government represents a necessary and important step in preventing suicide at a systems level. In response to the high rate of suicidal behaviors among Bhutanese refugees resettled in Vermont, the co-authors (PI: J.M.) began a suicide risk detection study in Burlington, aiming to deliver better risk detection and elucidation of culturally-responsive risk factors. The study was approved by the University of Vermont’s Institutional Review Board. In getting the project off the ground, we faced some challenges and learned a few useful lessons that we wish to share. First, in preparing the study materials (e.g., questionnaires, recruitment materials, scripts), we learned the importance of reserving enough time and money to go through a methodical, and well-organized, translation process. The process began by working with local bi-lingual cultural consultants to remove colloquialisms from measures and recruitment materials and ensure that English wording matched a comparable construct in Nepali. Subsequently, we partnered with a local translating firm to forward-translate the materials into Nepali and then hired bi-cultural and bi-lingual professional translators to back- translate the materials into English. Finally, discrepancies between forward-translations and back-translations were addressed in a consensus meeting. To illustrate the necessity of this process, when one of the measures we used was first translated from English to Nepali, the original translator misinterpreted the construct being tapped. The original wording of one of the questions on the Interpersonal Needs Questionnaire (Van Orden, Cukrowicz, Witte, & Joiner, 2012) was phrased as, “Other people care about me.” During the back-translation process, we learned that the original interpreter misconstrued the construct as “Other people take care of me.” This discrepant meaning was resolved in a consensus meeting. This entire translation process took nearly three months and cost approximately $1,000 for six questionnaires.

When beginning recruitment, we encountered a second challenge: Among many Bhutanese refugees, suicide is heavily stigmatized which, aside from reducing help-seeking behavior, has the practical effect of restricting the range of certain constructs of interest (i.e., suicide behaviors and risk factors). Few individuals, whether they have experienced suicide- related thoughts and behaviors or not, responded to recruitment flyers printed in both English and Nepali. Subsequently, we developed several outreach strategies including partnering with agencies that specialize in refugee, medical healthcare (Community Health Centers of Burlington and University of Vermont Medical Center’s New American Clinic), mental healthcare (Vermont Psychological Services’ Connecting Cultures specialty service, Community Health Centers of Burlington, and Networks Inc.), case management (Connecting Cultures, Howard Center, Association of African’s Living in Vermont) social work (Connecting Cultures and Howard Center), legal assistance (Vermont Law School, Vermont Legal Aid), and other local Bhutanese community associations. Through these partnerships, providers were trained to explain the nature and scope of the project and answer questions community members may have. A University of Vermont Institutional Review Board-approved referral process was established to ensure that this process was consistent with ethical and research guidelines. We wanted to ensure our process was not coercive or misleading. Since our project was not an intervention study, we wanted to make it explicit that participating in our study did not mean that participants would be receiving any sort of treatment as a part of our study. We provided scripts to all of our partnering outreach workers. As an aside, a referral list and local crisis number was distributed to every participant, and a safety plan was completed with each participant who endorsed any level of suicidality. If suicidal behavior was deemed imminent, procedures were clearly established to ensure participant safety. To minimize the risk of confidentiality breaches by outreach workers, we established clear guidelines that only referral partners who routinely comply with Health Insurance Portability and Accountability Act (HIPAA) and Family Educational Rights and Privacy Act (FERPA) laws would be allowed to make direct referrals. These referral partnerships led to increased interest in the study among Bhutanese refugee community members, and allowed us to each more community members struggling with suicide-related thoughts and behaviors, as evidenced by increased variability on suicide-related constructs.

A third practical lesson is that within the Bhutanese refugee community, many individuals require a Nepali interpreter for all in-person interactions with investigators and research staff, as well as for telephone correspondence. We partnered with local professional interpreters who indicted comfort in discussing issues related to suicide and suicidal behavior. We required our interpreters to complete training on HIPAA laws. As the study progressed, we gained an appreciation for the challenges that partnering with local interpreters presents. Although we gave participants their choice of individual interpreter, some participants still noted feeling as if their information was not protected when they worked with a local interpreter. The study team, subsequently, implemented a study-room set-up (i.e., a temporary opaque screen) so that the interpreter could remain blind to the participant’s identity (at least based on physical appearance), but the participant knows the identity of the interpreter. The success of this study and the meaningful results we hope it will provide have only been realized through active and strong community partnerships and a system in which community members are able to provide feedback about the study and adjustments that are needed to the research team. In line with these principles, we established a working Community Advisory Board of several community members to continually provide feedback to the study staff about how to improve the study, ensuring that it meets the needs of the local Bhutanese refugee community. We hope that the completion of this work will result in identifying culturally sensitive, accurate, and early risk factors for suicidal behavior to aid in detecting individuals within the Bhutanese refugee population at risk for suicide.

Public Significance Statement:

Bhutanese refugees have a suicide rate roughly twice that of the general U.S. population but a suicidal ideation rate that is roughly half that of the U.S. population. Proper culturally-responsive screening based on an ideation-to-action framework and a systems-based approach to prevention may be effective tools for reducing the number of suicides among Bhutanese refugees living in the United States.