Construct validity of a proposed new diagnostic entity: Acute Suicidal Affective Disturbance (ASAD)

Highlights

• Acute Suicidal Affective Disturbance (ASAD) is a unitary construct.

• Past ASAD symptoms predicted number of past suicide attempts.

• Past ASAD symptoms were highest in participants with multiple suicide attempters.

Abstract

Background

The current study presents initial support for the construct validity of Acute Suicidal Affective Disturbance (ASAD), a clinical entity consisting of acute suicide risk and several related features.

Methods

Participants (N=195) were university students who were recruited for a history of suicide attempt(s), history of suicidal ideation, or no history of suicide attempts or suicidal ideation. Participants completed study measures online.

Results

Factor analytic results indicated a one factor solution for a lifetime measure of ASAD symptoms. The measure demonstrated strong convergent and divergent validity with common correlates of suicide-related outcomes and incremental predictive validity, as lifetime occurrence of ASAD symptoms predicted number of past suicide attempts above and beyond a host of suicide risk factors. Lifetime ASAD symptoms differed between those with multiple suicide attempts, those with a single attempt, and participants without a history of attempts, as well as between participants with a history of both suicidal ideation and attempts and those with a history of suicidal ideation but not suicide attempts.

Limitations

The cross-sectional research design limits the ability to infer causation between ASAD symptoms and suicidal behavior. Only past ASAD symptoms (not current symptoms) were measured.

Conclusions

ASAD appears to be a unified clinical entity that characterizes acute suicide risk which may assist clinicians in determining a client’s potential for death by suicide.

Introduction

Suicide is a major global health concern, with over 800,000 dying by suicide annually (World Health Organization, 2014). Despite substantial research focusing on this phenomenon and increased prevention efforts, rates of death by suicide appear to be increasing (Centers for Disease Control and Prevention, 2013, World Health Organization, 2014). This alarming trend demonstrates the need for better characterization of imminent suicide risk, so that treatment and prevention may be tailored accordingly.

The majority of research on suicide has focused on risk factors and warning signs for suicide. Numerous variables (e.g., hopelessness, social isolation, non-suicidal self-injury) have been established as factors that increase risk of suicidal ideation and behavior (Beck et al., 1985, Nock et al., 2006, Van Orden et al., 2010); however, these risk factors poorly distinguish those who desire suicide from those who attempt it. Furthermore, most research uses suicidal ideation as the primary outcome variable; it may be that regularly cited risk factors for suicide are, in fact, risk factors for suicidal ideation (Klonsky and May, 2014). However, given that lifetime rates of suicidal ideation are many times higher than lifetime rates of suicide attempts (Kessler et al., 1999), Klonsky and May (2014) called for an ideation-to-action framework for understanding/predicting suicidal behavior, such that predictors and explanations for suicide should be classified as either addressing risk for suicidal ideation, suicide attempts among those with ideation, or both. Utilizing an ideation-to-action framework may improve models of suicide risk, as well as efforts to understand and prevent suicide (Klonsky and May, 2014).

There exists strong support for several variables as acute predictors of suicide attempts that can be integrated into an ideation-to-action framework. Suicide decedents often experience states of “severe and/or extreme agitation” prior to their deaths (Busch et al., 1993, Busch et al., 2003, Ribeiro et al., 2013, Robins, 1981). Sleep disturbances, particularly insomnia concerns, were also identified as acute indicators of lethal suicidal behavior in suicide decedents (Bernert et al., 2014, McGirr et al., 2007, Pigeon et al., 2012). When measured prospectively in a sample of psychiatric patients, agitation (i.e., “psychic anxiety”), insomnia, and the presence of panic attacks were significant predictors of suicide attempts within a 1-year follow-up period (Fawcett et al., 1990). Similarly, nightmares have been shown to be highly prevalent in adults presenting at medical units following a suicide attempt (Sjöström et al., 2007). Although this link remains unexamined, scholars theorize that feelings of disgust toward the self and others/the world may play an important role in suicidal behavior as this cognitive style likely influences the development of a host of important psychological predictors of suicide, such as perceptions of burdensomeness, self-criticism, and social alienation (Brunstein-Klomek et al., 2008, Chu et al., 2013, O’Connor and Noyce, 2008). This assertion is supported by research linking self-criticism to increased intent to die and lethality of suicide attempts in college students with a history of suicidal behavior (Fazaa and Page, 2003). Additionally, disgust was a very common feature among suicide decedents in Robins (1981) psychological autopsy study.

Existing theoretical models of suicidal behavior also provide insight into the characterization of imminent suicidal behavior. The interpersonal theory of suicide (Joiner, 2005, Van Orden et al., 2010) posits that marked social withdrawal represents a severe form of desire for suicide, characterized by perceived burdensomeness and thwarted belongingness. Perceived burdensomeness occurs when individuals feel that they are a liability on others, and that their deaths are worth more than their lives. Thwarted belongingness is characterized by feelings of loneliness and social disconnection. The interpersonal theory proposes that at high levels of the acquired capability for suicide, the combination of perceived burdensomeness and thwarted belongingness represents imminent suicide risk. Consistent with this hypothesis, indices of these three constructs predicted suicide attempt status (Joiner et al., 2009). Further, psychological autopsy studies have found that in the months, weeks, and days leading up to an individual’s death by suicide, suicide decedents are frequently described as becoming socially withdrawn and losing interest in social activities (Appleby et al., 1999, Chavan et al., 2008). Robins (1981) demonstrated that a third of suicide decedents were also described as being less talkative, and Phillips et al. (2002) showed that 29% of decedents experienced an interpersonal conflict in the two days prior to their deaths. Although not yet tested empirically, Ribeiro et al. (2013) proposed that marked social withdrawal following interpersonal conflict might explain the relationship between interpersonal conflict and death by suicide. Finally, research demonstrates that perceptions of problems being intractable long-term likely contribute to the development of suicidal behavior. Hopelessness has been shown to be a strong prospective predictor of suicide attempts (Kuo et al., 2004). Similarly, the interpersonal theory of suicide proposes that feelings of hopelessness regarding social alienation and perceptions of burdensomeness foster suicidal desire transitioning into suicidal intent (Van Orden et al., 2010). The integrated motivational–volitional (IMV; O’Connor, 2011) model of suicidal behavior posits that feelings of entrapment play a crucial role in the development of suicidal intent and suicidal behavior. In a study of inpatients hospitalized after a suicide attempt, entrapment was shown to predict subsequent hospitalization for one or more suicide attempts over a four-year follow-up (O'Connor et al., 2013).

In order to enhance the characterization of acute suicide risk, efforts need to be made to distill information from psychological states and theory-driven predictors of suicide that can inform clinical practice in a meaningful, practical way. A new clinical entity, Acute Suicidal Affective Disturbance (ASAD; Joiner et al., in preparation) may be valuable in explaining these relationships and in understanding imminent suicide risk. According to Joiner and colleagues, ASAD is characterized by the following: (a) geometric increase in suicidal intent over the course of hours or days (as opposed to weeks or months); (b) one (or both) of the following: marked social alienation (e.g., severe social withdrawal, disgust with others, perceptions that one is a liability on others), or marked self-alienation (e.g., views that one’s selfhood is a burden, self-disgust); and (c) perceptions that the foregoing are hopelessly intractable; and d) two or more manifestations of overarousal (i.e., agitation, insomnia, nightmares, and irritability). They further specify that the disturbance is not the result of an exacerbation of major depressive disorder, bipolar disorder, or substance use. This disturbance is theorized to be a unified construct, such that symptoms occur simultaneously alongside cardinal behavioral intent with a rapid onset. Much like depressive symptoms are a cluster of related symptoms subsumed under the same entity, we posit that the symptoms of ASAD similarly comprise a unified construct. Further, we suggest that ASAD may constitute a mental disorder according to the definition of the American Psychiatric Association, “A mental disorder is a clinically significant behavioral or psychological syndrome or pattern that occurs in an individual and that is associated with present distress or disability or with a significantly increased risk of suffering death, pain, disability, or an important loss of freedom” (APA, 2013). ASAD involves a clinically significant behavioral syndrome, is distressing, and quite certainly involves increased risk of death. The present study is one of multiple studies needed to establish a consensus definition of ASAD. Robins and Guze (1970) described a method for establishing diagnostic validity that consists of five phases: clinical description, laboratory study, exclusion of other disorders, follow-up study, and family study. The present study encompasses the first three phases.

Additionally, Joiner et al. (in preparation) theorize that this clinical entity is comprised of specific clinical factors that have been related to acute suicidal crises through empirical investigation, theory, and clinical accounts. Although constructs related to negative emotionality (i.e., negative affect, psychache, and emotional misery) are often related to suicidal thinking (O’Connor and Nock, 2014) and are likely highly related to ASAD symptoms, these constructs are less specific than the clinical characteristics of ASAD and are not as strongly linked to imminent suicidal behavior. Thus, negative emotionality is relevant to understanding suicide and likely highly related to ASAD criteria, but is not directly incorporated into ASAD due to their non-descript nature in relationship to acute suicidal behavior.

This new clinical entity will address major diagnostic issues that have been noted with regard to suicidal behavior. Specifically, suicide has been conceptualized as secondary to existing psychiatric diagnoses such as depression, leaving approximately 10% of those who die by suicide without a diagnosable mental disorder (Aleman and Denys, 2014, Oquendo et al., 2008). Relatedly, there are deaths by suicide that cannot be clearly attributed to an existing mental disorder (e.g., in scenarios in which the primary diagnosis is schizophrenia or anorexia nervosa), and yet, we view death by suicide as an exemplar of psychopathology; ASAD resolves this dissonance.

An expedient measure was proposed by DSM-5 to temporarily bridge the gap between the existing diagnostic void and a viable suicide disorder. Suicidal Behavior Disorder has been included under the ‘Conditions for Further Study’ section of the DSM-5 The criteria for this disorder were proposed primarily to facilitate assessment of suicidal behavior as an outcome, as opposed to criteria that would constitute a specific constellation of symptoms. Specifically, criteria do not reflect the experience of a suicidal crisis or imminent suicidal behavior, as ASAD criteria do, but rather indicate the presence of one or more past suicide attempts made with the intent to die within the last 24 months (current) or further back than 24 months (in-remission). The set of criteria for ASAD, on the other hand, proposes a disorder that would better reflect the type of valid clinical description that would meet the standards established by Robins and Guze (1970). For example, such a disorder would demonstrate discriminant validity from established mood disorders. A diagnostic entity that distills acute suicide risk factors in a meaningful way would aid clinicians in the understanding of who is most at-risk for imminent suicidal behavior beyond the clinician’s knowledge of other psychopathological features and how these features interact. Simply, a clinical entity like ASAD that has been derived by the extant literature of acute suicide risk prediction can aid in important clinical judgments’ regarding suicide risk without the reliance on the assessment and understanding of dozens of suicide risk factors, some related to acute suicidal crises and some not.

The purpose of the current study is to investigate the construct validity of ASAD. This clinical entity goes beyond simply an additional risk assessment instrument; rather, we propose that it is a mental disorder, and a distinct, essential, and dangerous one at that. With these intentions in mind, a measure of lifetime ASAD symptoms would have to demonstrate certain characteristics to provide initial construct validity for ASAD. First, a lifetime measure of ASAD should demonstrate a one-factor solution when subjected to factor analyses. Although ASAD consists of several related, but distinct, acute risk factors, these factors are theorized to cumulatively indicate acute suicide risk. In the current study, it was hypothesized that a one-factor solution would strongly fit the correlation matrix of a created measure of lifetime ASAD symptoms.

ASAD symptoms should also demonstrate differential relationships with psychological risk factors for suicide and the incidence of suicidal behavior. It was predicted that convergent/divergent correlations would demonstrate that lifetime symptoms of ASAD would be correlated, but not redundant, with psychological suicide risk factors (e.g., symptoms of depression and anxiety, suicidal ideation, and feelings of hopelessness). Lifetime ASAD symptoms should also be highly related to past suicidal behavior. It was hypothesized that the strongest correlations between lifetime ASAD symptoms and indicators of suicide risk would be found among those risk factors most closely related to suicidal behavior (e.g., non-suicidal self-injury and suicide attempts). It was also expected that lifetime ASAD symptoms would predict additional variance in the number of past suicide attempts above and beyond demographic and psychological predictors of suicidal ideation.

Further indication of the construct validity of ASAD and its theorized strong relationship to suicidal behavior would be demonstrated through the comparison of individuals with and without a history of suicide attempts. Lifetime ASAD symptoms should be elevated in individuals with a history of suicide attempts in comparison to those indicating a history absent of suicidal behavior, with highest levels of ASAD symptoms in those who have attempted suicide multiple times. It was expected that those with a history of multiple suicide attempts would demonstrate more lifetime symptoms of ASAD than those with only one past suicide attempt, as well as those without a history of attempting suicide. It was also hypothesized that those with both a history of suicidal ideation and past suicide attempts would endorse more lifetime symptoms of ASAD than those with a history of suicidal ideation alone.