The Narrative-crisis Model of Suicide as an Emerging Framework for Suicide Risk Stratification,Management,and Prevention

Suicide-related deaths in India have shown a consistent rise in the last few years, from 9.9 per one lakh population in 2017 to 12.4 per one lakh population in 2022. ¹ Notably, this phenomenon is nested in the background of a global waning in suicide rates in the last few decades, albeit before 2016. ² Further, time trends in India show a consistently high rate of suicide among students, a rise in suicides attributed to substance use, and a change in preferred suicide means, with deaths due to hanging recording a steady rise. ³ These observations, coupled with the continuing socioeconomic impact of the pandemic, provoke concern and call for action to identify factors underlying suicidal behaviors.

From a clinical perspective, a cornerstone of suicide prevention is the practice of suicide risk assessment and risk stratification. Suicide risk assessment is a highly structured process involving four main elements: assessment of risk factors; protective factors; specific suicide inquiry involving assessment of suicidal ideation, planning, and intent; and finally, assessment of evidence-based warning signs in suicide. This process is individualized and leads to stratified judgments about the level of suicide risk that inform subsequent management. ⁴

A key drawback of extant suicide risk assessment models is their reliance on chronic, longer-term risk factors to make judgments about the level of acute risk. This approach has issues because long-term risk factors for suicide, such as lifetime suicidal ideation, past suicide attempt history, and mental disorders, have not shown a satisfactory predictive ability for near-term individual suicidal behaviors. ⁵ One possible reason for this may be the static nature of these risk factors; because of this, they are more indicative of a chronic than acute risk of suicide. This biplanar distinction of suicide risk into chronic and acute risk, though infrequently done, has important clinical implications for practice. ⁴

A second, and potentially more significant, drawback is the excessive reliance on verbalized or elicited suicidal ideation (SI) as a gateway question for further assessments of suicide risk: if the patient denies SI when asked, it is recorded, and this line of questioning is abandoned. Such an approach is potentially tricky because self-reported ideation may be transient and inconsistent or never reported due to the client’s desire to conceal SI. Besides, cross-sectionally elicited SI has shown an inadequate predictive ability for near-term suicide behaviors. ⁶ Significantly, a retrospective chart review aimed at identifying dynamic risk factors among suicide decedents found that about two-thirds of them denied SI when last asked by a practitioner; half of these patients died within two days of the assessment. ⁷ These observations question the dependence on SI in risk assessment and highlight the need for a better understanding of drivers of individual progression from chronic to acute suicide risk.

In response to these gaps in the literature, several investigators have proposed multistage models of suicidal behavior that seek to explain the individual progression of suicidal states. Crucially, all of them distinguish between SI and eventual action. One such example is the interpersonal-psychological theory of suicidal behavior, ⁸ which posits that SI, resulting from feelings of thwarted belongingness and perceived burdensomeness, presages suicidal action, and is dependent on an acquired volitional capability for suicide; both ideation and action are necessary for suicide to occur. Likewise, the integrated motivational-volitional model ⁹ propounds that suicide occurs in two stages; the first stage is that of motivation, where SI emerges; in the next phase of volition, suicidal action occurs.

However, both the above models continue to rely on SI as a milestone in their understanding of suicidal risk progression. To address this potential limitation and to efficiently integrate long-term and near-term risk factors, investigators have proposed the narrative-crisis model of suicide (NCM). ¹⁰ Specifically, this model proposes that when a vulnerable individual with chronic risk factors experiences a triggering, stressful life event, it may trigger the development of a suicide-specific subacute, cognitive-affective state called the suicidal narrative (SN). The central feature of this state is an exaggerated negative view of self in relation to others. This perception leads to feelings of social defeat, humiliation, isolation (thwarted belongingness), and a subjective perception that one’s very existence is a burden on others (perceived burdensomeness). Given these considerations, suicide now becomes a viable option for the individual.

The evolution of SN triggers the development of the subsequent step in NCM: the suicide crisis syndrome (SCS), a mental state associated with a high near-term or imminent risk of suicide (days to weeks). The SCS is a more acute cognitive-affective state and involves five components in its latest formulation: entrapment (frantic state of cognitive hopelessness where the individual feels that they are stuck in a crisis with no avenues to solve or escape from the problem), affective dysregulation (emotional pain, panic attacks, dissociation, rapid mood swings, and acute anhedonia), hyperarousal (insomnia, agitation), social withdrawal (evading social contact and feelings of isolation), and cognitive dyscontrol (cognitive inflexibility, ruminative thoughts, and inability to control the thoughts). ¹¹

The SCS is proposed as a suicide-specific diagnosis and is currently being evaluated for inclusion in the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). ¹² Timely and accurate identification of SCS allows the delineation of a high-risk subgroup of individuals, regardless of overtly expressed SI, who need more intensive management and vigilance for suicide risk. Notably, expressed SI is not a milestone at any stage of the NCM, removing a potential limitation of previously described multistage models of suicidal behaviors.

Discussion

Usage of the full NCM in clinical practice depends on validating its components and the full model across cultures and settings. In this regard, both SCS and SN, assessed using the Suicide Crisis Inventory (SCI) and the Suicidal Narrative Inventory (SNI), respectively, have received considerable cross-cultural research attention. Studies examining the psychometric properties of SCI, both in general^{13, 14} and psychiatric populations,^{15, 16} have found evidence for a consistent factor structure, good reliability, and acceptable validity. Likewise, the SNI has demonstrated consistent and adequate psychometric properties.^{17, 18}

These findings have several important implications for our understanding and management of suicidal progression in individuals. First, a model with construct validity can be evaluated for the ultimate test in suicide prevention: its predictive validity for near-term suicidal behaviors. Such evidence is sparse, though Cohen and colleagues ¹⁰ have found encouraging results for the concurrent and predictive validity of NCM for prospective suicidal thoughts and behaviors (STB) in a diagnostically heterogeneous sample with mental health conditions at 1-month follow-up. However, the authors noted that the full NCM explained only 16.7% of the variation in prospective STB, indicating the role of other factors. These findings also align with the modest performance of other suicide prediction tools for near-term STBs.

Nonetheless, these findings point to acute changes in cognitive and affective state as potential drivers of transition from chronic to acute suicide risk. Interestingly, these domains are also central to another proposed acute suicide-specific syndrome: acute suicidal affective disturbance (ASAD). ¹⁹ The ASAD is characterized by a dramatic spike in suicide intent (hours or days as opposed to weeks or months), marked social alienation (active withdrawal, feelings of social isolation or disgust with others), or self-alienation (self-disgust or views that one’s existence is a burden to others), a hopeless outlook about the permanency of each of the thoughts and perceptions mentioned above, and symptoms of hyperarousal (irritability, agitation, insomnia). The existence of this syndrome may be construed as indicating discrete periods of elevated suicide risk or as a marker for recurrent increase in suicide risk.

An important advantage of NCM is that it provides a framework to inform precise interventions for suicide risk at different stages of an individual’s progression through the suicidal trajectory. At the starting point of this model are individuals with trait vulnerability for suicide, such as those with a history of childhood adversities, poor social support, insecure attachment, and personality traits such as impulsivity. These people are prone to develop overly negative cognitive representations of the self (the suicidal narrative) when experiencing a triggering, stressful life event (SLE). For individuals experiencing an SLE, interventions to enhance social support, coping, and problem-solving skills may be beneficial. ²⁰

Simultaneously, interventions targeting long-term risk factors that confer vulnerability for suicide, such as attachment therapy to heal negative attachment styles or dialectical behavior therapy (DBT) for personality deficits such as trait impulsivity and affective instability, may be indicated. Care providers should also target other predisposing and potentially modifiable risk factors, such as substance abuse, major psychiatric disorders, conduct problems, and a history of anger and violent behaviors, through appropriate pharmacological and psychosocial intervention strategies. Targeting these long-term risk factors may help to attenuate the risk elevation incurred by experiencing acute SLEs.

For those individuals who move to the next stage of NCM, namely, the suicidal narrative, interventions targeting the cognitive inflexibility of the negative self-narrative may be indicated. Some examples of suicide-specific cognitive interventions that are consistent with the principles of NCM and hence can be used to target the suicidal narrative are collaborative assessment and management of suicidality (CAMS), cognitive-behavioral therapy for suicide prevention (CBT-SP), and the attempted suicide short intervention program (ASSIP). ²⁰ Specifically, deficits in social support (for thwarted belongingness), cognitive restructuring (for fear of humiliation, defeat, and perceived burdensomeness), and goal setting (for goal disengagement and re-engagement) may be identified and targeted through these interventions.

The next most acute stage in NCM is the suicide crisis syndrome. In this stage, the client is likely to be in a state of heightened arousal and lability. Other symptoms include inflexible cognitive perceptions of entrapment, frantic hopelessness, and extreme anxiety associated with somatic symptoms, often bordering on panic. Given this near-psychotic presentation with acute difficulties in cognition, pharmacological interventions targeting potentially relevant neural circuits and neurotransmitter/autonomic systems would be indicated to attenuate suicide risk. For instance, the hyperarousal and panic-somatization symptom domains may respond to long-acting benzodiazepines, while cognitive rigidity may be targeted using small doses of antipsychotics. ²⁰

Other agents with proven anti-suicidal benefits, such as clozapine, lithium, and ketamine, may also be efficacious and helpful in the management of SCS. Once the patient is calmer and more amenable, psychotherapeutic interventions such as DBT to enhance skills related to emotion regulation, distress tolerance, and mindfulness may be considered for managing SCS. Figure 1 illustrates the risk mitigation intervention framework provided by NCM with proposed interventions at each stage of the model. Each stage of this framework remains to be empirically tested and provides significant research opportunities.

Figure 1.

DBT, dialectical behavior therapy.

On a cautionary note, some clinicians have raised concerns about the legal risks of using a suicide-specific diagnosis such as SCS and ASAD. ²¹ Specifically, the concerns surround the risks of lawsuits by plaintiffs alleging medical malpractice. However, in defense of suicide-specific diagnoses, one may argue that it provides several benefits that serve to mitigate the legal liability of the clinician. First, documentation of a suicide diagnosis assists clear communication during hand-offs and allows for consistent language between treatment teams; both reduce the possibility of oversight errors in critical areas such as patient vigilance.

Second, documentation of a suicide-specific diagnosis supports clinical practice and guards against it dropping below standards of care. For instance, documentation of a diagnosis of SCS would signal to the clinician that the patient requires an increased frequency of risk assessments, additional social support, and specific treatments such as safety planning. If a suicide-specific diagnosis and the corresponding treatment strategies are properly documented, the lawyer will surmise that the clinician has considered the possibility of suicide and taken appropriate interventions to address the risk of such an outcome. This is undoubtedly the best way to avoid malpractice lawsuits.

Conclusion

The narrative-crisis model of suicide is a multistage model that seeks to integrate long-term risk factors with near-term suicide-related variables. Its overarching goal is to explain the individual progression from chronic to acute suicidal thoughts and behaviors. The individual components of the NCM are validated widely, including in the Indian setting. Consequently, the NCM provides a theoretically comprehensive, empirically supported, and clinically meaningful framework for understanding, stratifying, managing, and preventing suicide. Future work must test the predictive validity of NCM for suicidal thoughts and behaviors and the efficacy of stage-wise interventions in preventing suicidal behaviors through randomized controlled designs. Therein, perhaps, lies its true test and utility.