Suicide ideation, planning, and attempts: the case of the Latinx LGB youth

Background: Guided by an ecological systems theory (EST) framework, the purpose of the present study was to investigate how multiple micro, mezzo, and macro factors influence the suicidality continuum from suicidal ideation to suicide attempt among Latinx LGB (lesbian, gay, and bisexual) youth living in the United States. Methods: Data for this cross sectional-study included 451 participants who self-identified as Latinx LGB on the 2017-National Youth Risk Behavioral Survey. The analysis explored micro, mezzo, and macro-level factors’ association with three suicidality outcomes (ideation, planning, and attempt) at the bivariate and multivariate level. Since the outcome variables were dichotomized, univariate logistic regressions and backward elimination logistic regressions were used. Results: The most commonly reported suicidal behavior was ideation (n = 173; 40%), followed by planning (n = 150; 34%), and then attempt (n = 64; 21%). Findings from the backward elimination logistic regression on suicidal ideation suggest the best set of independent variables are being bullied at school (odds ratio [OR] = 2.81; CI: 1.61–4.89), experiencing sexual assault(OR = 2.32; CI: 1.32–4.07), experiencing depressive symptoms (OR = 1.99; CI: 1.07– 3.69),being cannabis use (OR = 1.76; CI: 1.08–2.89), and being female (OR = 1.72; CI: 1.01–2.93).For suicide planning the model suggested, experiencing depressive symptoms (OR = 3.21; CI:1.74–5.91), cannabis use (OR = 2.46; CI: 1.49–4.07), being bullied at school (OR = 2.04; CI:1.17–3.58), and experiencing sexual assault (OR = 1.88; CI: 1.07–3.31) exhibited the strongest relationships. Suicide attempt was significantly associated with cannabis use (OR = 3.12; CI:1.60–6.08), experiencing depression (OR= 2.89; CI: 1.30–6.43), experiencing sexual assault (OR = 2.77; CI: 1.34–5.71), and being bullied at school (OR = 2.34; CI: 1.12–4.91). Conclusion: Given the findings of this study, it is essential that tailored suicide prevention efforts be established that uniquely address the intersections of race/ethnicity and sexual orientation and how this intersection influences micro, mezzo, and macro factors associated with suicide ideation, planning, and attempt among Latinx LGB adolescents.


Introduction
Despite the recognition that Latinx and sexual minority youth are groups that are at elevated risk of suicidal behaviors in the United States, little attention has been given to the experiences of Latinx youth who identify as lesbian, gay, and bisexual (LGB). This research oversight is sobering given that, separately, both groups are considered high-risk because they have some of the highest incidence rates of suicide. In 2017, a higher number of Latinx youth (8.2%) reported attempting suicide compared to non-Hispanic whites (6.1%). 1 Moreover, in 2017, 16.4% of Latinx youth seriously considered attempting suicide, 13.5% made a suicide plan, and 8.2% engaged in lethal action. A meta-analysis focused on sexual minority youth suggests that LGB youth are at increased risk of attempted suicide compared to heterosexual youth. 2 The same study revealed that relative to Latinx heterosexual youth, Latinx homosexual youth were 3.71 more likely to have attempted suicide, while bisexual youths were 3.69 more likely to have attempted suicide. Thus, suicide has now become one of the greatest threats to the health of Latinx LGB youth during the healthiest period of the lifespan. 3 For the purpose of this study, Latinx is a gender-neutral term describing youth living in the United States whose nationality group or the country in which the person or her/his parent's or descendants were born in is a Latin American country. 4,5 Using the ecological systems theory (EST), 6 the purpose of the present study was to investigate how multiple micro, mezzo, and macro factors influence the evolution HPP from suicidal ideation to suicide attempt among Latinx LGB youth. EST is a valuable paradigm because of the recognition of the strong association between individuals and various elements that make up their environments. EST underscores that a person's development and worldview is shaped by the transactional relationship with their social milieu. 6 According to the available literature, Latinx and LGB youth suffer from a number of micro-level stressors associated with suicidality. These micro-level stressors include cannabis use, [7][8][9] and experiencing: intimate partner violence, 10,11 sexual assault, [10][11][12][13] and depressive symptoms. [14][15][16][17] At the microlevel, sex differences have also been found in relation to suicide, with females experiencing higher levels of ideation than males. [18][19][20] At the mezzo level, identifying as a Latinx youth and having an LGB identity are risk factors in school settings. One study suggests that an association exists between identifying as a Latinx youth, experiencing bullying, and suicidal attempts. 21 Another study suggests that 70% of youth reported being bullied at school because of their sexual orientation, with 43% having been bullied on school property. 22 At the macro level, understanding the landscape of immigration-related policies are necessary since several studies suggest that immigration policies and the related exclusionary climate adversely affects the mental health of Latinx youth. [23][24][25] It is critical that more research is developed that focuses on investigating which socio-ecological factors promote increased notions of suicide among Latinx LGB youth because this group has some of the most elevated risks for suicide, which appears to disproportionately burden this group. To deal with this ethnic disparity, it is important to determine the pathways contributing to increased risk. Such knowledge is essential for developing a better understanding of the etiology of suicide among Latinx LGB youth, furthering risk assessment, and designing tailored evidence-based suicide prevention programs that address the particular needs of this population.
The following research questions guided the present study: 1. Controlling for sociodemographic differences, how do Latinx LGB youth compare and contrast across their reported suicide ideation, planning, and attempts? 2. What micro factors-such as depressive symptoms, cannabis use, IPV, and sexual assault-are associated with suicidality (i.e., suicide ideation, planning, and attempt) among Latinx LGB youth? 3. What mezzo factors-such as experiences at school with bullying-are associated with suicidality (i.e., suicide ideation, planning, and attempt) among Latinx youth? 4. Are macrosystem factors-such as state-level immigration-policy climate-associated with suicidality (i.e., suicide ideation, planning, and attempt) among Latinx LGB youth?

Study design and sample
This secondary data analysis used cross-sectional data from the 2017 National Youth Risk Behavioral Survey (YRBS). Using a three-stage cluster sampling design, the CDC has administered the YRBS annually since 1990 to monitor the health of 9th through 12th-grade students living in the United States. Data extracted for this crosssectional study represents a subsection of 2017 YRBS participants who self-identified as Latinx (N = 3653) and that also identified as LGB (N = 451). All demographic variables on the YRBS are weighted.

Instrumentation
The focus of this study was on parceling out suicidal behavior (i.e., ideation, planning, and attempt) among Latinx LGB youth. Thus, suicidal thoughts and behaviors were collapsed into three dependent variables: suicidal ideation; suicide planning; and suicide attempt. All three single-item variables were measured using responses to the following questions on the YRBS: During the last 12 months… "did you ever seriously consider attempting suicide?"; "did you make a plan about how you would attempt suicide?"; and, "how many times did you actually attempt suicide?". The latter was dichotomized into yes/ no response categories. If the participant responded that they carried out 1, or more, suicide attempts, they were coded as "yes". All response chooses were coded as 0 = no and 1 = yes.

EST domains
To represent the micro, mezzo, and macro-level domains, six EST variables were included in the analysis: cannabis use, depressive symptoms, intimate partner violence (IPV), sexual assault, bullied at school, and immigration policy climate. Cannabis use, depressive symptoms, IPV, sexual assault, and bullied at school were based on the participants' answers to the following YRBS questions: During the last 12 months… "how many times did you use marijuana?"; "did you feel so sad or hopeless almost every day for two weeks or more in a row that you stopped doing some usual activities?"; "how many times did someone you were dating or going out with physically hurt you on purpose?"; "how many times did anyone force you to do sexual things that you did not want to do?"; and, "have you ever been bullied on school property?". The response categories for marijuana use, depressive symptoms, sexual assault, and bullied at school were dichotomized into no (e.g., no endorsement of depressive symptoms or no reported incident of sexual assault) and yes (e.g., yes depressive symptoms were endorsed or yes reported at least one incident of sexual assault). Responses were coded as follows: 0 = no, and 1 = yes. For IPV, the response categories were collapsed into three categories: 0 = no IPV, 1 = yes IPV, and 2 = has never dated. Immigration climate was created using the participant's state of residency and Hatzenbuehler and colleagues' 25 14-item state policy 200 index. The policy index assigns a score representing the states immigration policy climate ranging from 1 (most inclusionary) to 5 (most exclusionary). We collapsed their categories into three: 1 = inclusionary, 2 = neither inclusionary nor exclusionary, and 3 = exclusionary. Lastly, participant's age (12 -14, 15, 16, 17, and ≥18) and sex (0 = male versus 1 = female) were also included in the analysis.

Statistical analysis
Univariate statistics were used to examine the variables' distributions and missingness, which revealed a large percentage of missingness in two variables, suicide attempt (31%) and sexual orientation (24%). This missingness was handled by first determining the nature of missingness using Little's MCAR test. 26 The statistically significant findings (P ≤ .001) from Little's MCAR test suggested that data was not missing-at-random; thus, commonly used imputation techniques were not appropriate to use, and instead, listwise deletion was used. The rest of the variables' missingness was below the acceptable percentage of 10%. After addressing missingness, a series of binary logistic regressions were conducted to explore each variables relationship to the three dependent variables (i.e., ideation, planning, and attempt) and the independent variables. A prior significance threshold of .25 or less was set for inclusion in the final multivariate analyses. 27 This final analysis included three multiple logistic regression models using backward elimination to find the set of best EST factors that of suicidality. In order to evaluate the models, four properties were examined: Akaike's information criterion (AIC), Bayesian information criterion (BIC), R 2 adjusted ( ADJ ), and Mallow's C p . The optimal model selected was one that exhibited: the smallest AIC, BIC, and Mallow's C p values; and, the largest R 2 ADJ value. 28 For the final multivariate analyses, statistical significance was measured at the 95% confidence interval level (P ≤ 0.05). All analyses were conducted using STATA version 12 (College Station, TX). 29 Odd ratios (OR) and 95% confidence intervals (CI) were reported for the bivariate and multivariate results. Table 1 shows the descriptive statistics of the variables used in the analysis. The Latinx LGB youth in our sample was predominantly female (n = 292) and 74% identified as bisexual with most youth between the age of 15 to 17 (n = 328). Many of the respondents reported not using cannabis (n = 275). Twenty-one percent of respondents reported experiencing depressive symptoms, while another 22% reported experiencing sexual assault. The majority of the sample also reported experiencing IPV (n = 237). Most of the respondents reported living in inclusionary states (n = 311).

Participant general characteristics
Nearly 40% of our sample reported experiencing suicide ideation, 34% reported making suicide plans, and almost 21% attempted suicide. Out of the 173 Latinx LGB youth who endorsed suicidal ideation in our sample, 71% of them also created a suicide plan and 34% of them attempted suicide. Similarly, of the 150 respondents who reported suicide planning 34% of them also reported having attempted suicide. All of the 64 Latinx LGB youth who reported having previously attempted suicide also reported a history of suicide ideation.

Predictors of suicidality variables among Latinx adolescents: univariate analysis
Results of logistical regressions showed that several micro and mezzo factors were significantly associated with each of the suicidality variables among Latinx LGB youth (see Table 2). Significant sex differences emerged with regards to suicide ideation, but not planning or attempt (OR: 1.22, 95% CI: .80-1.87; OR: .99, 95% CI: .54-1.81, respectively). Latinx LGB females had 1.69 higher odds of suicide ideation compared to Latinx LGB males. Significant age differences also emerged in to relation to suicide ideation and planning, but not attempts. Latinx LGB youth between the ages of 12-14 had .65 higher odds of suicide ideation than adolescents who were 15 years of age, while adolescents 16 years of age had .57 higher odds of reporting suicide ideation. However, Latinx LGB youth who were 16 years of age had .66 lower odds of suicidal planning than those in their early adolescence.
Among Latinx LGB youth, cannabis use, experiencing depressive symptoms, experiencing IPV, experiencing sexual assault and being bullied at school were all significantly associated with suicide ideation, planning, and attempt. Latinx LGB cannabis users had 2.12 higher odds of suicide ideation, 2.80 higher odds of planning, and 3.14 higher odds of attempting suicide than Latinx LGB non-users. Latinx LGB youth who experienced depressive symptoms had 3.67 higher odds of suicide ideation, 5.45 higher odds of planning, and 5.25 higher odds of attempting suicide than Latinx LGB with no reported depressive symptoms. Latinx LGB youth who reported having experienced sexual assault had 3.12 higher odds of suicide ideation, 3.01 higher odds of planning, and 3.78 higher odds of attempting suicide than Latinx LGB who reported not having experienced sexual assault. Latinx LGB youth who experienced being bullied at school were 3.23 times more likely to report suicide ideation, 3.61 time more likely to plan suicide, and 3.80 time more likely to attempt suicide than Latinx LGB who reported not experiencing bullying. In terms of IPV, Latinx LGB youth who reported engaging in non-violent intimate partner relationships were 4.49 more likely to report suicide ideation, 4.87 times more likely to plan suicide, but 8.56 time more likely to attempt suicide as compared to those who reported never experiencing IPV. Youth's specific sexual orientation (i.e., gay and lesbian versus bisexual) and the state-level immigration climate did not share a significant relationship with suicide ideation, planning, or attempt.

Predictors of suicidality variables among Latinx adolescents: multivariate analysis Suicidal ideation
The backward logistic regression model examining suicidal ideation among Latinx LGB youth suggested an optimal model that included five EST variables: sex, cannabis use, depressive symptoms, experiencing sexual assault, and being bullied at school (see Table 3). This model explained 11.67% of the variance in suicidal ideation outcomes among Latinx LGB. Latinx LGB females were nearly twice as more likely to report suicidal ideation as their male counterparts. Latinx LGB adolescents who experienced depressive symptoms were 1.99 more times likely to report suicide ideation compared to adolescents who had not experienced depressive symptoms, while adolescents who experienced sexual assault were 2.32 more likely to report suicide ideation compared to adolescents who had not experienced sexual assault. Latinx adolescents who reported cannabis use were 1.76 more likely to report suicide ideation compared to adolescents who were not cannabis users. The highest odds of reporting suicide ideation was among Latinx LGB youth who experienced bullying at school. They were almost three times more likely to report suicidal ideation as compared to those who reported no history of experiencing bullying.

Suicide planning
Findings based on the backwards logistic regression suggests the best set of independent variables of suicide planning among Latinx LGB youth included: cannabis use, depressive symptoms, experiencing sexual assault, and being bullied at school (see Table 3). This model with four EST factors explained 12.94% of the variance in suicidal planning. The Latinx youth in our sample who reported being bullied at school were 2.04 times more likely to report suicidal planning than those that reported not being bullied at school. Similarly, Latinx LGB youth who reported cannabis use were 2.46 times more likely to engage in suicide planning than those who reported no cannabis use. Latinx LGB youth experienced sexual assault were 1.88 times more likely to engage in suicide planning than those who did not experience sexual assault. The greatest odds of suicidal planning were amongst respondents who experienced depressive symptoms. Latinx youth who had depressive symptoms were three times more likely than those who reported no depressive symptoms to have made a suicide plan. LGB youth who had depressive symptoms were 2.89 times more likely to have attempted suicide than those with no depressive symptoms. Similarly, Latinx youth who experienced sexual assault were 2.77 times more likely to attempt suicide than those who did not experience sexual assault. Latinx LGB youth who were bullied at school had 2.34 times higher odds of attempting suicide than those who were not bullied. Respondents who used cannabis had the highest odds of attempting suicide; they were 3.12 times more likely to attempt suicide as compared to those who did not use cannabis.

Discussion
The results indicate that suicidality progress along the continuum from ideation to planning to attempt is a non-linear process among this population. Nearly half of the Latinx LGB youth who reported suicidal ideation did not report attempting suicide. More specifically, the results show that nearly 40% of Latinx LGB reported suicide ideation, another 34% developed a suicide plan, and nearly 21% attempted suicide. The prevalence rates of suicidal thoughts and behaviors reported in this study are higher than those reported by both groups independently in other studies. 1 These findings underscore the need for suicidality research to increase its focus on youth who hold multiple collective identities. The shared identities that intersect race/ethnicity and sexual orientation likely create a unique set of stressful circumstances that are driven by the system interaction of oppression, domination, and discrimination. 30 This systemic intersection constructs a social milieu that is constricting, often resulting in detrimental experiences for marginalized groups, such as Latinx LGB youth. Such context is important for health and mental health professionals to remember given that the development of an individual's identity and the capacity to function in everyday life is contingent on sociocultural and environmental processes. 31 Consistent with existing studies, suicide ideation among Latinx LGB youth was significantly associated with sex, [18][19][20] and experiencing depressive symptoms, 14-17 sexual assault, [10][11][12][13] and being bullied at school. 21,22 The findings corroborate other studies suggesting that significant sex differences exist with regards to suicide ideation, but not planning or attempt. 32 The findings show that the sex paradox of suicide behavior also exist among Latinx LGB youth. However, it is still unknown as to whether these gender differences are related to stigma in reporting ideation among boys, or because it is more socially acceptable and feminine for girls to be more cognizant of their emotional distress. 33 More specifically, research should explore the relationship between the multiple identities of ethnicity, sexual orientation, sex and gender.
Consistent with existing studies, 21,22,[34][35][36] the results indicate that being bullied at school is one of the consistent significant correlates of suicided ideation, planning, and attempt among Latinx LGB. Moreover, the results suggest that being bullied at school was the strongest factor associated with suicide ideation. These findings are very important given that school environments play a significant role in terms of development among youth that reaches far beyond academics, in large part due to their relationship with peers. The relationships in which they engage, both positive and negative, influence their mental health and social development. 37 Regrettably, for Latinx LGB youth, school is an unsafe space where they encounter damaging and degrading messages about not only their ethnic group, but also about their sexual orientation. 35 Considering the percentage of Latinx LGB youth who lack support from family and friends, 35 the degrading and damaging messages heard in school place Latinx LGB youth in a vulnerable position. It is often these experiences that propel Latinx LGB youth to develop feelings of insecurity, rejection, lack of self-esteem, depression, psychosomatic complaints, and feelings of shame associated with their identity. 38 Similar to other studies, [14][15][16][17] the results suggest that experiencing a mood disorder such as depressive symptoms is significantly associated with suicide ideation, planning, and attempt. This result is not surprising given that several of the harmful events experienced by Latinx LGB youth, such as lacking family support and acceptance, having unsupportive peer relationships, and experiencing negative social interactions often contribute to feelings of depression. 35,39 This finding is noteworthy considering that Latinx youth are disproportionately more likely to suffer from depressive symptoms than other racial and ethnic groups, 40,41 but they are also 53% less likely to have ever received mental health care 41 and they are less likely to receive treatment consistent with recommended guidelines. 42 As a result, mental health issues among Latinx youth often go overlooked and untreated.
Among Latinx LGB adolescents, the results of this study corroborate other studies suggesting that cannabis use is significantly associated with suicide ideation, planning, and attempt. [7][8][9] Moreover, in the present study, cannabis use was the strongest correlate of suicide attempt. This relationship may suggest a number of possibilities. It could imply that cannabis use impairs psychological and emotional functioning among Latinx LGB youth, which may bring about self-destructive actions that have tragic consequences, such as attempting suicide. However, this finding could also imply that cannabis use among Latinx LGB adolescents is a way of managing suicide thoughts and behaviors. 43 This explanation gives further credence to the self-medication hypothesis, which asserts that individuals use psychoactive drugs to lessen harmful emotional or cognitive states, such as suicide ideation, planning, and attempt. 44

Implications
Based on the results, several implications can be drawn. First, health professionals should recognize that factors that induce suicide ideation, planning, and attempt vary among Latinx LGB. For example, in some instances, Latinx LGB youth attempted suicide, but never ideated or planned it. Thus, each sequence of the suicide continuum should be treated independently. Second, depressive symptoms directly influenced most of the suicide stages, except for suicide attempt. Depressive symptoms appear to take a psychological toll on Latinx LGB youth, which may bring about more negative thoughts. Professionals who interface with Latinx LGB should take an inventory of their level of depressive symptoms. If depressive symptoms are high, practitioners should work to develop culturally-responsive intervention plans that include strategies to lessen depression but also promote psychological resilience. Similarly, health professionals should monitor levels of cannabis use among Latinx LGB youth since they were significantly associated with suicide ideation, planning, and risk and because Latinx youth in middle school are almost twice more likely to report cannabis use compared to non-Latinx White youth, 45 and LGB adolescents (50.4%) are more likely than non-LGB adolescents (28.8%) to report cannabis use. 46 This will be even more important today since more states in the United States have legalized recreational use of cannabis. Third, it is common for Latinx LGB youth to seek informal assistance from family, friends, and significant others but not formal providers. 47 Although this practice may serve youth well, this inclination should also be complemented by increasing professional outreach to this population. However, it should be noted that outreach efforts may not reach their full potential because of a shortage of diversity in the health and mental health workforce in the United States and a lack of contextually, culturally, and linguistically appropriate programs. There is growing recognition that identifying LGB counselors of color or programs that relate directly to their experiences with LGBTQ-and race-based discrimination is challenging. 22 Thus, initiatives that support cultivating a more inclusive workforce and programs that focus on the intersection of race/ethnicity and sexual orientation are especially needed. Fourth, there is a clear need to develop and implement anti-bullying federal and state policies protecting Latinx LGB students in schools. As of 2018, only 19 states and the District of Columbia have fully enumerated anti-bullying laws. 48 This suggests that more than half of the US does not protect Latinx LGB youth in schools from harassment, discrimination, and bigotry based on sexual orientation and/or gender identity.

Limitations
Despite the contributions made by the current study, several limitations should be acknowledged. First, the cross-sectional research design does not allow for any causal inferences. Second, the majority of the measures were single-items. Third, the variable "suicide attempt" had an unusually large percentage of missing data. Fourth, the dataset only included information about sexual identity (being gay, lesbian, and bisexual) but not other forms of sexual behavior or sexual attraction, which are other dimensions of sexual orientation. 49 Data on gender identity was limited. For example, Latinx youth were not given the options of self-identifying as queer, nor could they identify as transgender, or other non-binary gender identities. Last, the Latinx population is not a monolithic community; however, the diversity within this group was not captured by the analysis. There are several nationalities under the umbrella term of "Latinx". Unfortunately, the dataset did not allow for examining within-group differences that can highlight some of the differences that exist within this broad group. The dataset also did not account for differentiating between youth who were born in the United States versus those that had migrated to the United States.

Conclusion
Data from this study suggest that suicide ideation, planning, and attempt are highly prevalent among Latinx LGB adolescents. As a result, suicide has now become one of the single greatest health threats to a subset of the Latinx population, the largest racial/ethnic minority group in the United States. The results reveal that factors associated with each stage of suicidality vary significantly among Latinx LGB adolescents, which partially explains why the strongest factor associated with each model of suicidality differed. Overall, the independent factors examined in the present study did not account for a large variance in each regression model. Thus, using the EST framework to explain suicidality for this population is helpful, but more research is needed that explores situational and cultural factors of Latinx LGB youth and how these contextual and cultural experiences shape the suicide spectrum. Additionally, a better understanding and increasing awareness of the intersections of race/ ethnicity and sexual orientation and how this intersection influence suicidality may provide a pathway to lessen the burden of suicide Latinx LGB adolescents. The present study had a majority female sample, which highlights the need to develop further research that takes into account the experiences of cisgender Latinx males who are sexual minorities. Given the findings, it is essential that tailored suicide prevention campaigns be established that uniquely address micro, mezzo, and macro factors associated with suicide ideation, planning, and attempt among Latinx LGB adolescents. This is a call to action for families, peer groups, researchers, and practitioners to engage in the creation of novel interventions and messaging efforts that