# Violence Risk Assessment: Structured Professional Judgment

## Introduction

Assessing the risk of violence is among the most challenging and consequential tasks in clinical psychiatry. Unstructured clinical judgment alone performs poorly, while purely actuarial tools lack clinical flexibility. **Structured professional judgment (SPJ)** bridges these approaches by providing empirically grounded risk factors within a framework that allows for individualized clinical reasoning.

## Approaches to Violence Risk Assessment

### Unstructured Clinical Judgment

Based solely on the clinician's experience and intuition. Prone to cognitive biases: anchoring, availability heuristic, illusory correlation. Research consistently shows limited predictive validity (AUC approximately 0.55-0.65)

### Actuarial Assessment

Uses fixed algorithms derived from statistical models. Examples: **Violence Risk Appraisal Guide (VRAG)**, Classification of Violence Risk (COVR) Superior inter-rater reliability but inflexible to individual case context. Cannot incorporate dynamic or protective factors.

### Structured Professional Judgment

Combines empirically supported risk factors with clinical expertise. Clinician rates each factor and forms a final risk judgment (low, moderate, high) Allows integration of **static, dynamic, and protective factors**. Considered the current best-practice standard.

![Comparison of unstructured, actuarial, and SPJ violence risk assessment approaches](images/violence-risk-approaches.png)

## Key SPJ Instruments

| Tool | Setting | Items | Unique Feature | Outcomes Assessed |
|------|---------|-------|----------------|-------------------|
| HCR-20 V3 | Forensic, civil, correctional | 20 (10H + 5C + 5R) | Most widely used; historical + dynamic + management factors | Violence |
| START | Inpatient, forensic | 20 | Assesses both vulnerabilities AND strengths | Violence, self-harm, substance use, victimization, elopement |
| SARA | IPV assessment | 20 | Specialized for intimate partner violence | Spousal assault risk |
| VRAG-R | Forensic (actuarial) | 12 | Purely statistical; no clinical judgment | Violent recidivism |
| PCL-R | Forensic | 20 | Gold standard for psychopathy assessment | Informs violence risk (not risk tool per se) |

### HCR-20 (Version 3)

The most widely used SPJ tool internationally. **Historical factors** (10 items): prior violence, young age at first violent incident, relationship instability, substance use problems, major mental illness, personality disorder, traumatic experiences, violent attitudes, treatment noncompliance, employment problems. **Clinical factors** (5 items): lack of insight, negative attitudes, active symptoms, instability, treatment or supervision non-response. **Risk management factors** (5 items): professional services, living situation, personal support, treatment compliance, stress or coping.

### Short-Term Assessment of Risk and Treatability (START)

Designed for inpatient and forensic settings. Uniquely assesses both **vulnerability factors and strength factors** across 20 items. Generates risk estimates for multiple adverse outcomes: violence, self-harm, substance abuse, victimization, unauthorized leave.

### Spousal Assault Risk Assessment (SARA)

Specialized tool for intimate partner violence risk. 20 items covering criminal history, psychosocial adjustment, and spousal assault history.

## Risk Factors for Violence

### Static Risk Factors

History of prior violence (the single strongest predictor) Young age at first violent incident. Male sex. History of childhood abuse or witnessing domestic violence. Antisocial personality traits.

### Dynamic Risk Factors

Active psychosis, particularly **threat-control override symptoms** (delusions that external forces are threatening or controlling the individual) Substance intoxication or withdrawal. Medication non-adherence. Acute psychosocial stressors. Command auditory hallucinations (modest independent risk)

### Protective Factors

Engagement in treatment and therapeutic alliance. Stable housing and employment. Prosocial support networks. Absence of substance use. Effective coping skills and insight into illness.

![Risk factor domains in the HCR-20 V3 structured professional judgment tool](images/hcr20-risk-domains.png)

## Communicating Risk

Avoid categorical statements such as "this patient is dangerous" or "this patient is safe". Frame risk as **contextual, dynamic, and conditional**: "Under the following conditions, the risk of violence is elevated..". Specify the time frame, nature of potential violence, and likely targets. Recommend risk management interventions tailored to identified factors.

## Documentation and Legal Considerations

Document the risk factors assessed, the tool used, and the clinical reasoning behind the final judgment. Risk assessment is a **process, not a one-time event**; reassess when clinical circumstances change. Know your jurisdiction's duty-to-protect laws (Tarasoff and its progeny) Failure to perform a structured risk assessment may constitute a deviation from the standard of care.

![Template for documenting a structured violence risk assessment](images/violence-risk-documentation-template.png)

## Key Clinical Pearls

Mental illness alone is a weak predictor of violence; substance use and psychopathy are far stronger risk factors. The base rate of violence in the general population is low, which means even good tools will produce many false positives. SPJ is not about predicting violence with certainty but about identifying modifiable risk factors to guide management. Always assess risk in the context of potential targets, available means, and environmental constraints. Reassessment should occur at every transition of care: admission, discharge, privilege changes, and after critical incidents.

## References

1. Douglas KS, Hart SD, Webster CD, Belfrage H. *HCR-20V3: Assessing Risk for Violence*. Mental Health, Law, and Policy Institute, Simon Fraser University; 2013.
2. Singh JP, Grann M, Fazel S. A comparative study of violence risk assessment tools: a systematic review and metaregression analysis of 68 studies involving 25,980 participants. *Clin Psychol Rev*. 2011;31(3):499-513.
3. Webster CD, Martin ML, Brink J, Nicholls TL, Desmarais SL. *Manual for the Short-Term Assessment of Risk and Treatability (START)*. Version 1.1. BC Mental Health and Addiction Services; 2009.
4. Monahan J, Steadman HJ, Silver E, et al. *Rethinking Risk Assessment: The MacArthur Study of Mental Disorder and Violence*. Oxford University Press; 2001.