Prevention-Oriented Suicide Risk Assessment: Planning Not Prediction

May 05, 2016

News Type:  Director's Corner
Author:  Anthony Pisani, PhD, Associate Professor of Psychiatry and Pediatrics, Center for the Study and Prevention of Suicide at University of Rochester

Most clinicians-in-training learn to summarize suicide risk in a categorical probability judgment expressed as low, moderate, or high, often with gradations like low-moderate. But what do we really mean when we say a patient is at “low” “moderate” or “high” risk? Risk compared to whom? Compared to when? In what setting? These labels are devoid of context, lack predictive validity, and provide little help when developing plans and responses to prevent suicide.

In an article recently published in Academic Psychiatry, my colleagues Daniel Murrie and Morton Silverman and I suggest an alternative model of risk formulation to guide clinical practice and workforce education. We propose a shift from prediction-oriented to prevention-oriented thinking and language. Despite exciting research progress occurring in the field to improve short-term risk prediction, we are a long way from being able to apply this research to specific decisions about individual patients at a given time point. And even if we get better at short-term prediction, we need models that directly inform prevention plans and management.

A focus on prevention means anchoring risk formulation in the specific context of a patient’s own history and the clinical setting in which the assessment occurs. With regard to the patient’s own history, risk formulation should capture the fluid nature of suicide risk and explicitly state how the person’s current risk compares to risk at previous time points and how risk might change in response to future events. With regard to clinical setting, a patient considered high risk in one context (e.g., a college counseling center) might be considered low risk in another context (e.g., an inpatient psychiatric hospital). These risk appraisals differ, not only because patient populations differ, but also because each setting has different resources available for intervention. Likewise, the purpose of an assessment varies by setting. So, clinicians must conceptualize and describe risk in relative terms. Describing a patient as low risk or high risk in the abstract is far less meaningful than describing the patient as at lower or higher risk relative to other patients in the same context.  

We define risk formulation as a concise synthesis of evidence-based suicide risk information regarding a patient’s immediate distress and available resources at a specific time and place. Building on the important work of many colleagues in suicide and violence risk assessment, we propose a prevention-oriented, contextually anchored model for synthesizing data into four distinct judgments to directly inform intervention plans:
 
1.    Risk status – The patient’s risk relative to a specified subpopulation
2.    Risk state – The patient’s risk compared to baseline or other specified time points
3.    Available resources – Resources from which the patient can draw in crisis
4.    Foreseeable changes – Changes that may exacerbate risk

Understanding risk in context, identifying resources available to the patient, and anticipating events that might increase (or decrease) a patient’s risk empowers clinicians, patients, and families to develop specific plans for using resources, heading off potential triggers, and creating a path to recovery.

In our experience, the best communication and planning occurs when individual clinicians or entire clinical teams/systems share a framework and language (not just a set of forms) for assessing and managing suicide risk. Furthermore, medical anthropologists tell us that shared language and frameworks shape cultures of care—a key element of the aspirational goal of Zero Suicide.

Much work remains to validate new models and test their impact on patient care and outcomes, but it is clear that current assessment language is lacking. The time is right to move beyond outdated frameworks, including categorical risk predictions, and move toward new models that are anchored in the context of patients’ lives (past, present and future) and the clinical settings where they are seen. The goal of risk formulation is not to categorize or predict but to plan collaboratively toward life-saving preventive actions.

For more information on this topic, see “Reformulating Suicide Risk Formulation: From Prediction to Prevention” by Anthony Pisani, Daniel Murrie, and Morton Silverman, an article which originally appeared in Academic Psychiatry and is available to read or download at no cost from the SPRC Library.