Evaluating the Interpersonal-Psychological Theory of Suicide among Latina Adolescents using Qualitative Comparative Analysis

Abstract

The Interpersonal-Psychological Theory of Suicide (IPTS) has emerged as an empirically supported theory of suicide risk, yet few studies have utilized IPTS to examine the suicidal behaviors of Latina adolescents. In this study, we explore the cultural and developmental appropriateness, as well as the explanatory fit, of IPTS within a sample of Latina adolescents. Data for this project were drawn from qualitative interviews conducted with Latina adolescents with (n=30) and without (n=30) histories of attempted suicide. We employed a deductive qualitative approach to define and compare core constructs of IPTS (perceived burdensomeness, thwarted belongingness, and acquired capability), and then use qualitative comparative analysis to evaluate how core constructs were linked with the occurrence of a suicide attempt. Consistent with IPTS, perceived burdensomeness, thwarted belongingness, and acquired capability were present in 20 of the 30 adolescents who had attempted suicide, and absent in 22 of the 30 adolescents with no lifetime history of suicidal behaviors. Notably, alternative combinations of IPTS constructs were found in 10 cases of adolescents who attempted suicide, suggesting a need to adjust IPTS to fit the developmental and cultural contexts of Latina teens. Although our results suggest predominantly positive support for IPTS, participants varied in terms of how their experiences resonated with the conceptual definitions put forward by the theory. Ultimately, our findings point to the ways in which developmental tensions are exacerbated by broader sociocultural dynamics, contributing to a broader understanding of suicide risk among ethnic minority adolescents.

In 2015, Latina adolescents had the highest rates of suicide attempts (15.1%) when compared to their Asian-American (11.1%), African-American (10.2%), and non-Hispanic White (9.8%) female peers (Centers for Disease Control and Prevention [CDC], 2016). Elevated rates of suicidal behaviors among Latinas were first reported in the early 1960s by clinicians serving Puerto Rican patients in South Bronx, New York, (Trautman, 1961a, 1961b). In the decades that followed those initial reports, studies with community-based and clinical samples of Latina adolescents provided additional empirical evidence of this mental health disparity (Friedman, Asis, Boeck, & Difiore, 1987; Garofalo, Wolf, Wissow, Woods, & Goodman, 1999; Ng, 1996; Razin et al., 1991; Rew, Thomas, Horner, Resnick, & Beuhring, 2001; Tortolero & Roberts, 2001). Substantial confirmation of the scope of the disparity arrived with the launching of the CDC’s Youth Risk Behaviors Surveillance (YRBS). In 1995, 21% of Latina adolescents participating in the YRBS reported attempting suicide, compared to 10.8% of Blacks, and 10.4% of White female adolescents (CDC, 1996).¹ Successive YRBS data have continued to show a persistent pattern of elevated risk for suicidal behaviors among Latina adolescents (Romero, Edwards, Bauman, & Ritter, 2014).

Previous efforts to explain the disparity in suicidal behaviors among Latina adolescents have applied an ecodevelopmental framework, which proposes that suicidal behaviors are shaped by adolescent developmental factors and family dynamics, on one hand, and sociocultural processes, on the other hand (Author, 2005). Through the application of an ecodevelopmental approach, researchers have found that Latina adolescents who attempt suicide often grow up in families with elevated levels of interpersonal conflict between caregivers and teens, and adolescent developmental processes play an integral role within this conflictual environment. Specifically, different sociocultural perceptions around issues related to autonomy strain the development of supportive interpersonal relationships within the family. Such conditions shape an adolescent’s vulnerability to emotional stress and influence the onset of a suicide attempt (Author 2012, 2015; Romero et al., 2014).

Although an ecodevelopmental approach brings much-needed attention to the interaction between individual (e.g., adolescent development), proximal (e.g., familial dynamics), and distal (e.g., adolescent rearing norms) factors shaping the suicidal behaviors of Latina teens, this model lacks conceptual specificity to elucidate how various individual-level risk factors come together to culminate in a suicide attempt. To this end, the Interpersonal-Psychological Theory of Suicide (IPTS; Joiner, 2005; Van Orden, Witte, Gordon, Bender, & Joiner, 2010) has emerged as an empirically-supported theory of suicide risk that might shed light on how risk factors, such as family conflict, influence suicidal behaviors at the individual level. In this study, we explore the cultural and developmental appropriateness, as well as the explanatory fit, of IPTS within a sample of Latina adolescents who attempted suicide. In doing so, we hope to contribute to a broader understanding of suicide risk within adolescent populations.

The Interpersonal-Psychological Theory of Suicide

IPTS suggests that suicidality emerges at the intersection of an individual’s desire to die by suicide and the ability to carry out that desire (Joiner, Van Orden, Witte, Selby, Ribeiro, Lewis, & Rudd, 2009; Joiner, 2005). The theory posits that two psychological states contribute to the desire to die: “perceived burdensomeness” and “thwarted belongingness.” Perceived burdensomeness originates in feelings about being ineffective or lacking self-efficacy, resulting in negative cognitive appraisals about one’s self-competence (Van Orden, Lynam, Hollar, & Joiner, 2006). Over time, individuals come to view themselves as so incompetent that they perceive: 1) themselves to be an inordinate burden to others; and 2) that the world would be a better place if they died (Van Orden et al., 2006; Van Orden, Cukrowicz, Witte, & Joiner, 2012). Under such apperceptions, a desire for suicide can emerge. Thwarted belongingness is an individual’s perception of not belonging or being socially isolated. Belonging refers to a basic psychological need for social connection, which is crucial for well-being (Baumeister & Leary, 1995). When the need for belonging remains unfulfilled, or “thwarted,” the emotionally painful consequences of social isolation can bring about or enflame a desire for death (Van Orden et al., 2012; Whitlock Wyman, & Moore, 2014).

IPTS proposes that suicidal behaviors occur when an individual develops a desire for death alongside an acquired capability for suicide. The theory suggests that repetitive exposure to painful events desensitizes individuals to pain and reduces the fear of death (Joiner et al., 2009). As Joiner argues, an acquired capability for suicide develops “when self-injury and other dangerous experiences become unthreatening and mundane—when people work up to the act of death by suicide by getting used to its threat and danger” (2005, p. 48). In general, acquired capability is hypothesized to emerge via chronic exposure to any event that provokes a pain response or fear of death (Wolford-Clevenger et al., 2015). Put another way, individuals can develop a capability for suicide over time through experiences that diminish their fear and sensitization to pain (Gautier et al. 2014). Non-suicidal self-injury (Germain & Hooley, 2012; Guan, Fox, & Prinstein, 2012), physical and sexual abuse (Joiner et al., 2007; Smith, Monteith, Rozek. & Meuret, 2017), and witnessing or participating in acts of violence (Jordan & Samuelson, 2016; Wolford-Clevenger et al., 2015) have been suggested as potential experiences that give rise to such desensitization.

Since the development of IPTS, numerous studies have supported the validity of the theory to advance an understanding of suicidal behaviors (Chu et al., 2017). Yet, within this body of research, investigators have relied primarily on clinical and community samples with adults (Van Orden, Cukrowicz, Witte, & Joiner Jr, 2012) and individuals who self-identify as White (Stewart, Eaddy, Horton, Hughes, & Kennard, 2017). To consider the developmental and cultural specificity of IPTS, there is a need to examine whether the constructs of perceived burdensomeness, thwarted belongingness, and acquired capability function similarly; and to evaluate the explanatory potential of the theory with adolescents from diverse ethnic backgrounds (Ribeiro & Joiner, 2009; Suh et al., 2017). To date, few studies have utilized IPTS to examine the suicidal behaviors of Latina adolescents (Kene, Brabeck, Kelly, & DiCicco, 2016). Such a limitation is noteworthy given that Latina teens are a growing population at heightened risk for suicidal behaviors (Croyle, 2007; Price & Kuhchandani, 2017; Romero et al., 2014).²

The limited existing research on how IPTS applies to diverse demographic and cultural populations makes qualitative methods an especially appropriate fit for addressing how Latina youth experience the constructs of IPTS (Sandelowski, 2000), and how these constructs might help explain suicidal behaviors (Ragin, 1999). Although most qualitative researchers work inductively with their data to develop concepts and theories from the ground up (Marecek, 2003); qualitative methods can also contribute significantly to the testing, refinement, and revision of theoretical models through deductive qualitative research (Gilgun, 2005; LeCompte & Schensul, 2012). Such an approach begins with an initial conceptual model, such as IPTS, that is evaluated, modified, and revised through the process of qualitative analysis (Bernard, Wutich, & Ryan, 2017; Camic, Rhodes, & Yardley, 2003). In this article, we utilize a deductive approach to define and compare perceived burdensomeness, thwarted belongingness, and acquired capability within a sample of Latina adolescents who have attempted suicide. Additionally, we evaluate the theory’s explanatory value by assessing whether IPTS can be used to potentially explain the suicide attempts of Latina teens when compared with peers with no lifetime history of suicidal behaviors.

Methods

In this study, we analyze a sub-sample of qualitative data collected from a larger mixed-methods project that explored suicidal behaviors among adolescent Latinas in low-income families in New York City. In the qualitative phase of the study, 139 adolescents were referred for participation by clinicians who assessed their self-harming behaviors as suicidal. The participants included 73 Latinas aged 11 to 19 who had self-harmed within six months prior to the interview and 66 Latina adolescents with no reported lifetime history of self-harm. Potential participants were excluded if they had a diagnosis of bipolar disorder, schizophrenia, psychotic disorder, or intellectual disability because these conditions present unique challenges that might uniquely shape the etiology of suicidal behaviors (Hauser, Galling, & Correll, 2013; Fujita et al., 2015; Martin, Thomas, Andrews, Hasking, & Scott, 2015; Mayes, Calhoun, Baweja, & Mahr, 2015). As described elsewhere (Author 2010, 2015), participants who self-harmed were recruited from outpatient and inpatient departments associated with three different hospitals and outpatient mental health services in a local community agency serving a large Latino population. Participants with no history of self-harm were recruited from primary care medical clinics and community agencies providing youth programming. The research team sought and received IRB approval at all institutions involved in project activities, and all participants (adolescents and their caregivers) provided assent and consent for participation in the study.

Below, we present data from a subsample of participants in the larger study. Among participants with histories of self-harm, we selected for analysis only those adolescents who explicitly stated in the qualitative interview that the intention of her actions was to die. From the total sample of 73 participants, 37 disclosed a suicide attempt with the intent to die. Participants were then matched by key demographic variables to adolescents with no histories of self-harm. Seven participants could not be matched due to differences in age, and these interviews were excluded from the analysis. This led to a final sub-sample of 60 participants (30 who had attempted suicide, and 30 with no lifetime history of suicidal behavior) matched by age and place of birth. Of those participants, over 60% were also matched by legal status and Hispanic sub-group.

In the sub-sample, the average age of the girls was 16 years. The majority were born in the United States (60%), and participants identified with seven Hispanic subgroups: Colombian, Dominican, Ecuadoran, Mexican, Puerto Rican, Salvadoran, and Venezuelan (see Table 1).

Table 1.

Characteristics of Adolescent Latinas in Sub-Sample

	Participants who attempted suicide (N=30)		Participants with no history of attempted suicide (N=30)		Total participants (N=60)
		n (%)		n (%)		n (%)
Age in Years	16.0 ± 2.2		15.6 ± 2.2		16.0 ± 2.2
11–13		7 (23.3)		7 (23.3)		14 (23.3)
14–16		14 (46.7)		14 (46.7)		28 (46.7)
17–19		9 (30.0)		9 (30.0)		18 (30.0)
US Born		18 (60.0)		18 (60.0)		36 (60.0)
Legal Status
Citizen		19 (63.3)		22 (73.3)		41 (68.3)
Resident		9 (30.0)		6 (20.0)		15 (25.0)
Undocumented		2 (6.7)		2 (6.7)		4 (6.7)
Hispanic Cultural Group
Puerto Rican		7 (23.3)		10 (33.3.)		17 (28.3)
Dominican		11 (36.7)		9 (30.0)		20 (33.3)
Mexican		4 (13.3)		4 (13.3)		8 (13.3)
Colombian		3 (10.0)		5 (16.7)		7 (11.7)
Other1		5 (16.7)		2 (6.7)		6 (10.0)

Notes:

¹This includes participants who identified as Ecuadoran, Salvadoran, Venezuelan, and mixed

Qualitative Interview

In-depth, open-ended interviews were conducted with all research participants in Spanish or English, depending on the participant’s preference. All interviewers were bilingual Latinas with masters- or doctoral-level training in a mental health field (e.g., social work, psychology). Each interviewer was trained to conduct qualitative interviews to encourage participants to talk freely, openly, and in-depth about topics in the interview guide. When applicable, the participants were asked to provide a retrospective and detailed discussion of the suicide attempt. For all participants, the interview guide explored perceptions of family and social life, including relationships with family members, peers, and romantic relationships, and past-experiences of victimization and suicidal and non-suicidal self-harm. All interviews were digitally recorded and transcribed in the original language of the interview. Qualitative interviews lasted between 25 and 70 minutes.

Data Analysis

To achieve our aims, we developed a three-stage strategy for deductive analysis to categorize and contextualize our data (Maxwell & Miller, 2008). Stage 1 focused on evaluating the cultural and developmental appropriateness of IPTS by examining how each IPTS construct manifested within the narratives of Latina teens. In Stage 2, we evaluated whether IPTS could be used to explain the occurrence of suicidal behaviors in Latina teens. In Stage 3, we integrated our analyses from Stage 1 and Stage 2 to contextualize those cases that did and did not fit the theory. We outline each of these stages in more detail below.

Stage 1.

First, we deductively developed three codes (perceived burdensomeness, thwarted belongingness, and acquired capability) and their code definitions based on the literature. We then compared, revised, and contextualized each of these codes within participant interviews (Padgett, 2016). To accomplish this, the first and third author read two interview transcripts independently, comparing similarities and differences in the behaviors, experiences, emotions, and meanings associated with the three broad codes. Specifically, we compared and contrasted code definitions based on the literature while remaining open to new ways in which the codes could manifest based on the perspectives and experiences of participants.

During a team meeting, we discussed our perspectives and interpretations of the data, and revised the codebook to document nuances based on data emerging in the interviews. This process continued with subsequent interviews until no new changes were made to the codebook, which occurred after reading 24 interviews. In the finalized codebook, each code (perceived burdensomeness, thwarted belongingness, and acquired capability) was accompanied by a series of sub-codes that reflected both the theoretical construct as described in the literature and the meanings and experiences of that code as narrated by participants (see Table 2). Sub-codes helped to reveal not only the different meanings associated with each IPTS code, but also the context within which such a code might emerge. With the final codebook constructed, all interviews were double-coded by the first and third author. Additionally, all coding was independently evaluated by the second author. The purpose of this evaluation was to verify the application of each code and to deepen the analysis by including multiple interpretations of the data set (Tuckett, 2005). All coding discrepancies were discussed during a team meeting, and coding was revised after achieving consensus.

Table 2.

Codebook: Definitions, Experiences, and Contexts of IPTS Constructs

Code	Definition	Experiences	Example from the Interview
Perceived Burdensomeness	Perception that one is incompetent or lacking self-efficacy (Van Orden et al., 2012) Feeling one is a burden to others (Van Orden et al., 2006) Feeling that one’s life circumstances are a burden (data-driven definition)	Worthlessness Feeling like a failure Hopelessness	“Life is pointless. Like, if I’m not gonna be anything, like, I don’t want to be just like another piece of scum on earth. I’d just rather die. I wouldn’t have to wake up anymore. I wouldn’t have to worry about how am I gonna graduate from high school, what I have to do to get into a college, all the work, trying.”
Thwarted Belongingness	Social isolation, or an unmet need for social connection (Van Orden et al., 2008)	Avoiding social connection with others Marginalized or alienated by others Changes in one’s social network (e.g., parental death, immigration)	“I have nobody I can talk to. I’m in a country that is not mine. And I don’t know anything. So, [in my home country], I know everything. I have my grandparents. But here, it’s a country, like, wow, really strange for me.”
Acquired Capability	Exposure to or direct experiences of physically painful events (Joiner, 2005)	Exposure to violence in the home Exposure to violence in school or community Experiences of physical and/or sexual abuse Self-directed harm	“I just get so angry. And I want to do something about it, but I can’t. I wanted to get this rage that I have out of me. So. I started to hurt myself. I cut myself.”
Context	References the context within which IPTS constructs could emerge (data-driven definition)	Mental health challenges Family struggles Academic difficulties Major life transition Discrimination	“You wouldn’t see happiness in my house. I think my mom cheated on my dad. She made my dad miserable. That’s when my father started doing more drugs, started drinking more, and he left the house.”

Stage 2.

In Stage 2, we used qualitative comparative analysis (Ragin, 1999) to examine how codes patterned in different combinations across the data. Qualitative comparative analysis (QCA) begins with the premise that complex phenomena, such as suicidal behavior, cannot be reduced to a single causal explanation (Ragin, 1999). The purpose in QCA is to analyze how different configurations of codes relate to the presence or absence of a phenomenon. In doing so, QCA reveals potential pathways that give rise to that phenomenon. QCA employs the principles of Boolean algebra to examine patterns of codes that exist in dichotomous states, such as present or absent (Bernard, 2013). Because QCA works with codes in dichotomous form, the approach has been described as exemplifying many of the characteristics of quantitative research (Rihoux, 2006). However, as Rantala and Hellström point out, “there is nothing quantitative in the underlying aim…QCA analyses are simply useful phases in the ongoing interpretative process of trying to categorize and understand data” (Rantala & Hellström, 2001, p. 89).

In our application of QCA, we transformed the coded data into binary form (present, absent). To accomplish this, we first organized all coded data into a conceptual cluster matrix (Miles & Huberman, 1994). In the matrix, data were organized by rows (participants) and columns (codes; perceived burdensomeness, thwarted belongingness, acquired capability). Individual cells in the matrix contained summarized data comprised of sub-codes, brief examples of coded text, and participant quotes. For example, the cell for a participant who expressed perceived burdensomeness would include summarized text of the experience and context of the burdensomeness, alongside an illustrative quote. All text was accompanied by a citation that identified its location in the interview transcript (page and line number), thus retaining a close link with the original data.

Following the construction of the conceptual cluster matrix, data in each cell were reduced further in a separate matrix to reflect either the presence or absence of each code across all participants. We then organized the reduced matrix into a QCA table, or “truth table,” that listed all potential code combinations, given their presence or absence (Bernard, 2013, p. 492–499; Haworth-Hoeppner, 2000; Ragin, 1999). We analyzed the truth table to evaluate how certain patterns, such as the presence of perceived burdensomeness, thwarted belongingness, and acquired capability, were linked to the occurrence of a suicide attempt in our sub-sample. We also identified negative cases, or outliers, that did not seem to support IPTS.

Stage 3.

In Stage 3, we utilized an iterative process of working across the conceptual cluster matrix, the truth table, and individual interview transcripts to contextualize our emergent findings. First, we grouped participants according to their specific configuration of codes. For example, all participants who expressed perceived burdensomeness, thwarted belongingness, but not acquired capability, were grouped together in the conceptual cluster matrix. This was facilitated by using the “sort” and “filter” features in the spreadsheet software. Then, we analyzed similarities and differences in each group using details in the conceptual cluster matrix, alongside additional readings of interview transcripts, to gain insight and contextualize why specific cases did or did not fit the theory. We discussed our interpretations of each case in a team meeting to limit the potential for bias and promote alternative explanations (Ryan & Bernard, 2000).

Results

In the results described below, we explore the cultural and developmental appropriateness of IPTS for Latina adolescents. First, we focus on how adolescents defined and interpreted their experiences as related to the core IPTS constructs of perceived burdensomeness, thwarted belongingness, and acquired capability. We describe how each IPTS construct manifested in the narratives of adolescent participants who had attempted suicide and illustrate the various circumstances that gave rise to perceived burdensomeness, thwarted belongingness, and acquired capability. Second, we discuss how different configurations of IPTS constructs were coded across participants with and without lifetime histories of suicidal behaviors to evaluate the explanatory potential of IPTS in a sample of Latina adolescents. All names used below are pseudonyms.

Cultural and Developmental Appropriateness of IPTS Constructs

Perceived Burdensomeness.

According to IPTS, perceived burdensomeness stems from perceptions of self-incompetence (Van Orden et al., 2012). Over time, individuals develop the sense that their ineptitude is so severe that it becomes a burden to other individuals in their social network, and that suicide provides a way to alleviate that burden (Van Orden et al., 2006).

Among adolescents in our sample who attempted suicide, every participant described feeling incompetent or inept (n=30). Several different circumstances shaped these self-perceptions, such as experiences of parental criticism or academic struggles. In some cases, for example, participants internalized harsh parental criticism. As Aimee recalled, “my mom, she curses, telling me that I’m dumb. That kind of things.” Such criticism resonated with Aimee, and led to her to conclude that she was “worthless.” Other participants in our sample described the high standards set for them in terms of academic success, such as achieving good grades and getting into college. Such expectations were sometimes beyond what some girls felt they could realistically accomplish. For example, Daniela, who overdosed on her mother’s prescription medication, reported feeling significant levels of pressure to succeed academically. She attributed her poor academic performance, and her family’s reaction, as shaping her experiences of perceived burdensomeness. As she explained,

“It’s just so much pressure for me to do so much. I know my father didn’t mean to make me feel like a failure, but that’s what he made me feel. Like, because he was saying, ‘You should have worked harder.’ It’s just like a constant feeling of failure with everything.”

Despite the circumstances that engendered feelings of incompetence, only a few adolescents in our sample reported that they attempted suicide to alleviate a perceived burdensomeness to others. As Leticia explained, “I felt like I was not doing anything in this life. I wasn’t helping anybody. So I drank the bleach and decided to take my life.” In cases like Leticia’s, participants feel that their families would be happier without them. In the words of Lisa, “I can’t live with this uncertainty, anxiety, and fear. So I’m going to disappear, to see if (the family) can be happy for the first time in their lives.”

In contrast, most participants in our sample experienced perceived burdensomeness somewhat differently. Rather than feeling they were a burden to others, participants felt that their life circumstances were a burden to them. For example, Isabel described protracted conflict with her mother, which ultimately led her to live with her grandparents. On the day of her suicide attempt, she found herself reflecting on her relationship with her mother and others in her social network. As she recalled, “I was like ‘What was the whole point of me living when I was gonna keep getting hurt?” From the perspectives of our participants, life was characterized by suffering, and there was no hope for change in the future. The burden of the emotional pain was felt to be, in the words of Daniela, “too much.” Unable to identify ways to ameliorate their pain, a desire for suicide emerged.

Thwarted Belongingness.

Thwarted belongingness, or social isolation, references an individual’s unmet need to establish a social connection with others (Van Orden et al., 2008). Within our sample, 26 adolescents who attempted suicide reported experiences coded as thwarted belongingness. Participants in our sample attributed their thwarted belongingness to three distinct sources: some saw it as the result of broader changes to their social networks, while others saw it as emerging from their own coping strategies or experiences of alienation within their family and peer networks.

The most common source of thwarted belongingness stemmed from changes in participants’ social networks. The death of a family member, parental divorce, immigration, or a change in school could usher in a sense of loneliness and isolation. For example, Lizbeth noted that immigrating to the United States made it difficult to develop friendships due to language barriers. She noted, “I couldn’t get friends because they speak English. I don’t have anybody to talk to, who can support me. And I feel sad.” Lizbeth linked her suicide attempt directly to her experience of thwarted belongingness, explaining, “Nobody supports me. I have no one to talk to. So, I just want to die so I don’t have to be with all of this.”

Other participants noted that despite the availability of social support from specific family members or friends, they were reluctant to emotionally and socially engage with others. As Vanessa commented, “I always keep my feelings inside. Nobody knows if I’m mad or sad. Like, I don’t want nobody to know what I feel at that moment.” Similarly, Daniela expressed, “I just have a problem communicating with other people. That’s just the way I’ve always been. I just, I’m used to keep everything to myself. I’m used to not going to others. Just dealing with everything by myself.” The decision to keep one’s internal experiences private, thus avoiding the development of a social connection with others, frequently stemmed from a desire to avoid another person’s reaction. For example, Jessica had described being in treatment for depression prior to her suicide attempt, yet she would not talk with her friends about what she was experiencing. As she reflected, “Like, they try to like cheer me up or whatever. But, they really weren’t. Like, I would just lie about it, after all. I’ll be like, ‘Yeah, I’m okay,’ when I really wasn’t. I didn’t want people to bother me about it.”

Another source of thwarted belongingness was feeling marginalized or alienated by others, including parents, family members, or peers. Veronica, for example, described an estranged relationship with her mother. “I used to be telling her that I loved her, or she loves me. She never answered. She didn’t know what to answer.” While most often thwarted belongingness was experienced within family relationships, some participants described experiences of discrimination in their broader social spheres. Vanessa noted that a teacher called her a “spick,” and she explained, “I did not like that. I had a conflict with that teacher because who is she to make fun of my race? Because I’m a Hispanic, it gives her a right to call me a ‘spick?’”

It is important to note that the three sources of thwarted belongingness did not occur in isolation. Rather, participants could experience multiple sources at the same time. For example, Sofía had recently immigrated to the United States. Prior, she had been living with her mother in Colombia, as her father had immigrated to the United States four years earlier. As she explained, “my mother raised me.” When Sofía was 14, her mother, who had been suffering from kidney disease, died from complications following a kidney transplant. As she remembers, “The transplant was a success. And everything was fine. But the kidney became infected, and the infection spread throughout her body. Then, they couldn’t do anything for her.” Following the death of her mother, Sofía joined her father in the United States. Although her father “was there” to provide support, she felt disconnected from all that she had known in Colombia. Filled with sadness and pain, Sofía recalled, “I didn’t feel like talking with him. Because [my mother’s death] affected him, too. When I would start crying and tell him, ‘I miss my mom,’ he would get sad. So, I kept it all inside me.” Sofía described “hurting” immensely without her mother, and as a result, she attempted suicide. Reflecting on that day, she noted, “I thought it was the best way to fix my problem.”

Acquired Capability.

According to IPTS, an acquired capability for suicide emerges over time via exposure to physically painful events (Joiner, 2005). Among participants in our sample, twenty-four adolescents described an acquired capability for suicide. The most prominent source of acquired capability was exposure to violence, including witnessing intimate partner violence between caregivers in the home, direct experiences of physical or sexual abuse, or being involved in physical fights in school. Additionally, some adolescents reported histories of self-harm. Often, girls reported exposure to multiple sources of acquired capability and their narratives suggested that these experiences were interconnected.

For example, Julia, who had experienced violence in her home environment, provided an example of how different sources of acquired capability could be interconnected. She noted that her father used to “beat” her mother. In response, her mother encouraged Julia to protect herself. As Julia explained, “My mom, she said, ‘Always defend yourself. Not matter what you have in your hand. I don’t care who it is who bothers you. Just go ahead and smack them in the face.’” In the fourth grade, Julia’s peers began to taunt her about her weight, and she responded physically. When asked to describe one of the fights, Julia said, “I didn’t use no weapons. I used my hands. And me and him, we got into an argument, and I made him bleed and stuff.” Julia noted that she had been in three physical fights, in total, that school year.

While Julia’s example shows how the socialization to violence can result in the girl’s engagement in violent acts with others, Mia’s case illustrates how experiences of physical abuse can shape adolescents’ decisions to self-harm. As Mia shared, she was six years old when her cousin began to abuse her sexually. The abuse continued for nearly three years, until Mia and her family moved to a different state. When Mia was eight, she began to cut herself. As she remembered, “I had the knife. And I stabbed myself, like that, in my arm. And in my leg. I have a mark.”

Histories of engaging in self-harming behaviors (non-suicidal self-injury and previous suicidal behaviors) were important sources of acquired capability in our sample. Carla, for example, started harming herself when she was 11 years old. “Like I would just, like, punch a wall, or like cut myself a little bit where [my dad] couldn’t see. Kind of stuff like that.” After cutting herself, she would irritate the wound to prevent it from healing. She described that “I just kept picking it. I just kept hurting myself. And I liked it. It helped with my anxiety.” Other girls reported previous suicide attempts. Romina recalled that when her parents were getting divorced, she felt immense “feelings of sadness. Disappointment. With everything that was coming down on me.” The first time she attempted suicide, she tried to hang herself. As she reported, “I put the rope around my neck and fell asleep. When my mom came in, I was purple. She was screaming, and I was sleeping. I was unconscious.”

Explaining Suicide Attempts Using IPTS: Qualitative Comparative Analysis

To evaluate the explanatory potential of IPTS, Table 3 describes each possible logical combination of perceived burdensomeness, thwarted belongingness, and acquired capability given their presence or absence, and how each combination manifested across the study sample of adolescents with and without histories of suicidal behaviors. Consistent with IPTS, each construct was present in 20 of the 30 adolescents who had attempted suicide. Notably, each IPTS construct was also present in two adolescents with no lifetime reported history of suicidal behaviors but who expressed suicidal ideation. For example, one adolescent, Zoe, said that “I might as well end it.” Zoe reported feeling worthless and that she was “losing everybody.” Moreover, she described a history of cutting, sexual assault, and fighting at school. Considering the expressions of suicidal ideation, and presence of each IPTS construct, this suggests a potential risk for progression from suicidal ideation to attempted suicide in these two cases.

Table 3.

Qualitative Comparative Analysis of IPTS Constructs across Cases (Truth Table)

Perceived Burdensomeness	Thwarted Belongingness	Acquired Capability	Suicide Attempt (n=30)	No History of Attempted Suicide (n=30)
yes	yes	yes	20	2
yes	yes	no	6	2
yes	no	yes	4	1
no	no	no	0	22
no	yes	no	0	3
no	no	yes	0	0
yes	no	no	0	0
no	yes	yes	0	0

In line with IPTS theory, each construct of perceived burdensomeness, thwarted belongingness, and acquired capability was absent in 22 of the 30 adolescents with no lifetime history of suicidal behaviors. It is important to note, however, that alternative combinations of IPTS constructs were found in 10 cases of adolescents with histories of suicidal behavior, as shown in Table 3. Each of these are explained below.

Six participants who had attempted suicide did not have evidence of acquired capability. In each of these cases, the girls described experiencing a sudden major life transition. Five of the six participants had immigrated to the United States, which ushered in many changes for which the girls felt unprepared, particularly in the breakdown of the social networks they had established in their countries of origin. Additionally, two girls experienced the death of their mother. It is probable that under such conditions, the intensity of the emotional pain combined with a lack of social connection, may have been enough to shape an adolescent’s decision to attempt suicide. For the girls who lost their mothers, experiencing the death of a parent may have eliminated the fear of dying. For example, Maria explained that at the time of her suicide attempt,

“I just felt really depressed. I was really missing my mom… It was, it’s a feeling that, it’s almost non-explainable. I just felt, I felt very cloudy. I felt like you see in the cartoons when it’s just like one cloud on them. I felt like it was just raining on me.”

Additionally, four participants who attempted suicide expressed perceived burdensomeness and acquired capability, but not thwarted belongingness. Each adolescent in these cases reported having a diagnosis of depression and receiving mental health treatment at some point. Although they recognized and acknowledged a strong social network, the emotional burden of depression was felt to be too much. As Amelia explained, “They were all like negative thoughts. Like, I’m not good enough. I’m not smart. I’ll never be anything successful.” When asked why she attempted suicide, she noted, “I was really depressed.”

Discussion

Our results indicate predominantly positive support for IPTS as a theory that can shed light on suicidal behaviors in Latina adolescents. Unique to our study is the testing of IPTS with a sample of Latina adolescents. Given that most research of IPTS has been conducted with adult populations (Van Orden et al, 2012), and predominantly White samples (Stewert et al., 2017), our approach extends previous work using a sample of younger and ethnic minority study participants. While our results suggest the potential application of IPTS within our sample, it is important to highlight the importance of adjusting theoretical concepts to fit cultural and developmental contexts.

For example, we found that among Latina adolescents, perceived burdensomeness had additional dimensions than those described in the literature. For some participants, their descriptions and experiences fit the definition suggested IPTS. For the majority of participants, however, perceived burdensomeness included the feeling that the world was a burden to them. This finding might be explained using developmental and cultural frameworks.

Adolescent development and acculturation are simultaneous and interdependent processes for Latina adolescents (Author, 2016). As Latina adolescents develop, they must confront contrasting cultural expectations (Calzada, Fernandez, & Cortes, 2010). On one hand, many young Latinas must adapt to Latino cultural expectations for girls to care for the family. Studies have shown that Latina adolescents report making material and emotional sacrifices for the good of the family, such as spending less money on themselves or suppressing negative feelings that might thwart interpersonal harmony (Esparza & Sánchez, 2008; Fuligni, Teseng, & Lam, 1999; Gonzalez, Stein, Shannonhouse, & Prinstein, 2012; Schwartz, 2007). Furthermore, some Latina teens who attempt suicide have been shown to describe their suicide as a sacrifice that would alleviate strain on the family unit or prevent future individual failures that would impact the family (Author, 2012).

On the other hand, Latina teens are also socialized within a mainstream US culture that often privileges becoming independent from one’s family of origin. Navigating this process may create a sense of inadequacy and burden for adolescents as they negotiate the conflicting developmental expectations for adolescent girls in both cultures. The way our participants experienced and reflected perceived burdensomeness might reflect how normative developmental tensions and conflicts are exacerbated by cultural conflict. We offer this preliminary interpretation as an arena for future research on IPTS.

Additionally, it is important to note that we identified several negative cases whose experiences of suicidal behavior did not align with IPTS, reminding us to remain cognizant of the fact that the experience of suicidality is not “one size fits all.” For example, immigration played a unique role in shaping thwarted belongingness among our participants. Girls who were born outside the U.S. described how the process of immigration disrupted their social networks with family members living in their countries of origin, and how language barriers continued to isolate them from making friends in the U.S. Five out of six girls who attempted suicide and did not present with acquired capability described immigration to the U.S. as a major life transition. This suggests that the stressful and ongoing consequences of immigration might shape suicidal behaviors in the absence of acquired capability or evoke intense emotional pain that can escalate a youth’s potential for suicide (Klonsky & May, 2015).

Ultimately, our study helps to shed light on the role played by individual-level risk factors in the suicidal behaviors of Latina adolescents, while at the same time, providing relative empirical support to IPTS. In doing so, we hope to contribute to a broader understanding of suicide risk to advance empirically-supported prevention and treatment practices for ethnic minority adolescents.

Limitations

It is important to acknowledge the limitations of our study. Participants in our study were recruited from the New York City area, and the generalizability of our findings to Latina youth living in other regions of the United States cannot be assumed. Additionally, we did not compare gender differences as such an approach would require the incorporation of individuals across gendered identities. Research has suggested gender differences in the experiences of the constructs of perceived burdensomeness, thwarted belonginess, and acquired capability and their links to suicidal behaviors. For example, Czyz and colleagues (2015) found a significant interaction between thwarted belonginess and acquired capability among adolescent girls only (Czyz, Berona, & King, 2015). Such findings resonate with the broader theoretical literature suggesting that adolescent girls are socialized to be more sensitive to and cognizant of social relationships (Cyranowski, Frank, Young, & Shear, 2000). Moreover, we do not include a cultural comparison group (e.g. non-Hispanic Whites, African Americans, Asian Americans), which would elucidate the interaction of cultural processes with IPTS constructs. The process of unpacking gendered and cultural differences in risk variables can contribute rich nuances and suggest how suicidal behaviors can best be prevented and/or addressed for youth across diverse identities. Finally, we also acknowledge this study’s focus on risk versus youth resiliency. As described by IPTS (Joiner, 2005; Van Orden et al., 2010), connectedness and belonging may be points for prevention and intervention within Latino populations.

Despite these limitations, the use of a deductive qualitative approach has provided important insights into how the concepts of IPTS can be situated within the narratives of Latina adolescents. While quantitative studies help to identify predictors and mediators for suicide, qualitative studies are particularly useful in exploring how theories and interventions can be adapted for minority populations and specific age groups (Rosen et al., 2017). Additionally, the use of in-depth qualitative interviews as units of analysis can extend preexisting knowledge on suicidality among Latina teens and demonstrate the nuances of the Latina teenage experience that cannot be captured by quantitative data.

Conclusion

Overall, study results support the potential application of IPTS to understanding individual-level risk factors that predict suicidal behavior among Latina teenagers. However, scholars must be attuned to how the IPTS constructs fit within the context of a Latina teen living in the U.S. today. Future research is needed to test the applicability of IPTS to other samples of Latinas adolescents and minority groups. Also, future studies should utilize a resiliency approach to guide an exploration of why individuals do not attempt suicide despite the presence of risk factors. Finally, this study supports the use of deductive qualitative research methods to understand the subtleties of suicidality and to bridge the gaps between theory and its application to effective real-world interventions.