Do styles of emotion dysregulation differentiate adolescents engaging in non-suicidal self-injury from those attempting suicide?

Highlights

• While there is agreement that emotion dysregulation acts as a risk factor for harmful behaviors (e.g., suicide, non-suicidal self-injury[NSSI]), there is a lack of consensus about what features define it and how best to assess it.

• We looked at how different assessments of emotion dysregulation might differentiate mutually exclusive groups of adolescents: (1) those who have made a suicide attempt (SA), with no history of NSSI (SA-only); (2) those who have engaged in NSSI, with no history of an SA (NSSI-only); (3) community-recruited, typically developing controls without any form of psychopathology including NSSI or SA.

• Mean comparisons suggest that adolescents with a history of NSSI reported significantly lower distress tolerance and higher emotional reactivity when compared to adolescents who made an SA.

• After controlling for shared variance across emotion dysregulation measures, parent report of affective lability was the only scale to uniquely distinguish between NSSI and SA groups.

Abstract

Emotion dysregulation is implicated in both suicide attempts (SA) and non-suicidal self-injury (NSSI). However, little is known about how emotion dysregulation may differ between adolescents who have made an SA from those engaged in NSSI. We sought to address this gap by comparing emotion dysregulation profiles across three homogenous groups of adolescents (1) SA-only (2) NSSI-only (3) and typically developing controls (TDCs). Mean comparisons suggest that adolescents with a history of NSSI reported significantly lower distress tolerance and higher emotional reactivity when compared to adolescents who made an SA. After controlling for shared variance across emotion dysregulation measures, parent report of affective lability was the only scale to uniquely distinguish between NSSI and SA groups. Accurately distinguishing emotion dysregulation patterns across self-injurious groups has practical implications towards assessment, treatment, course of illness, and prevention.

Introduction

Suicide is currently the second leading cause of death among 13–18-year olds. Yet, completed suicides are just one part of the larger public health crisis, with even more children making suicide attempts (SA) or having recurrent suicidal ideation (SI) (Centers for Disease Control and Prevention [CDCP], 2016). That is, according to the 2017 Youth Risk Behavior Survey, during the year prior, 17.2% of high school students nationwide seriously contemplated suicide, 13.6% made a suicide plan, 7.4% made an SA, and 2.4% made an SA that required medical treatment (CDCP, 2018). Despite continued policy, research, and clinical efforts, suicide trends have remained largely unchanged for decades (Kann et al., 2016; Kann et al., 2018).

Non-suicidal self-injury (NSSI) is another form of self-harm defined as deliberately hurting one's own body tissue, dissimilar to SAs, without intent to die (O'Carroll et al., 1996). Epidemiological reports have shown alarming lifetime rates of NSSI among adolescents, ranging from 6% to 30% in community samples, whereas only 4–6% of adults endorse the behavior (Klonsky, 2011; Plener et al., 2009; Muehlenkamp et al., 2012). Adolescence may be a critical period for studying NSSI when considering its rates peak at this age and mortality risk is higher for teens compared to young adults (Jacobson and Gould, 2007; Olfson et al., 2018). While there are clear differences between NSSI and an SA—including frequency, prevalence, and function—there are also substantial areas of overlap. Approximately 40% of adolescents who self-injure have also made a lifetime SA (Whitlock and Knox, 2007), and NSSI is regarded as a robust predictor for future suicide behaviors (Asarnow et al., 2011; Klonsky et al., 2013; Wilkinson et al., 2011). In fact, NSSI provides a tenfold risk for future SA, even after accounting for previous SA and depressive symptoms (Wilkinson et al., 2011).

Unfortunately, most prior studies have used samples that engage in both NSSI and SA, perhaps not surprising given high rates of co-occurrence. For example, while one study compared typically developing controls (TDC) with a group of individuals having made an SA, they inconsistently included those who had also engage in NSSI (Nock and Kessler, 2006). Another study compared TDC with no SIB to those engaging in NSSI and to those with both NSSI+SA, yet no group of individuals with SA exclusive of NSSI (Brausch andGutierrez, 2010). There remains a great need to clarify factors that are specifically and uniquely related to NSSI vs. SA. The work of Stewart et al. (2017), comparing three groups of adolescents engaging in NSSI (i.e., non-ideators, ideators with no lifetime SA, ideators with lifetime SA), contributes to this shift in the field. They found that those with a lifetime attempt engaged in more NSSI methods and in more severe forms of NSSI (e.g., burning) compared to non-ideators. Continued focus is needed on disentangling the mechanisms underlying each form of SIB, such as emotion dysregulation, requiring comparative studies involving homogenous groups of people engaged in NSSI-only or SA-only, but not necessarily both NSSI+SA.

Emotion dysregulation is a transdiagnostic factor, cutting across psychopathology categories and implemented in the development, maintenance and exacerbation of child/adolescent psychopathology, including internalizing, externalizing and personality disorders (e.g., Aldao et al., 2010; Cicchetti et al., 1995; Glenn and Klonsky, 2009; Morris et al., 2010; Neumann et al., 2009; Valiente et al., 2007). Emotion dysregulation is more specifically implicated in both forms of self-injurious behaviors (SIB), clearly supported by the influx of research linking the construct as a precursor to NSSI, SA (Adrian et al., 2011; Rajappa et al., 2012) and other maladaptive and impulsive behaviors (McLaughlin et al., 2011). While there is agreement that emotion dysregulation acts as a risk factor for harmful behaviors, there is a lack of consensus about what features define it and how best to assess it. As a result, there are competing models over what these factors consist of, and how to measure them (Cole et al., 2004; Zeman et al., 2006). This is mirrored in the number of scales adopted to measure aspects of emotion dysregulation. In general, these models refer to poor understanding and awareness, low acceptance, and maladaptive management of emotions (Gratz and Roemer, 2004; Mennin et al., 2007). Previous studies analyzing the link between emotion dysregulation and self-injurious behaviors are limited because they only focus on a single assessment measure, or a few components of these constructs. For example, the Difficulties in Emotion Regulation Scale (DERS) indicated that one dimension, “beliefs about access to regulation strategies”, accounts for a greater variance in NSSI behaviors compared to other aspects of emotion dysregulation measured on the DERS (Perez et al., 2012).

The current study sought to probe how different assessments of emotion dysregulation might differentiate mutually exclusive groups of adolescents: (1) those who have made an SA, with no history of NSSI (SA-only); (2) those who have engaged in NSSI, with no history of an SA (NSSI-only); (3) community-recruited, TDC without any form of psychopathology including NSSI or SA. Specifically, we evaluated between-group differences on six measures of emotion dysregulation. Then, we tested whether specific dimensions of emotion dysregulation predicted group status by accounting for shared overlap across scales. We hypothesized that both SA and NSSI participants would have more severe emotion dysregulation profiles vs. TDCs. However, a priori hypotheses of how the NSSI and SA groups would differ were not set due to the novel design of directly comparing mutually exclusive NSSI and SA groups on multiple features of emotion dysregulation