VRAG: Violence Risk Appraisal Guide
Violence Risk Appraisal Guide · Also known as: VRAG, Harris-Rice-Quinsey VRAG
The Violence Risk Appraisal Guide (VRAG) is an actuarial instrument developed by Harris, Rice, and Quinsey (1993) to estimate the probability of violent recidivism among adult male offenders released from forensic psychiatric hospitals. It represents one of the earliest empirically validated violence prediction tools and remains widely used in correctional, forensic psychiatric, and civil commitment settings. The VRAG combines static historical variables with dynamic factors to generate a probability-based risk assessment.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
The VRAG is designed specifically for adult males released from forensic psychiatric hospitals or forensic custody settings. It is most appropriate for post-discharge risk assessment, parole/conditional release decisions, and civil commitment hearings involving individuals with a history of serious violence and psychiatric disorder. The instrument has been validated in Canadian, American, and international correctional and forensic psychiatric populations. It is not suitable for youth, females, or individuals without substantial prior contact with the criminal justice system or psychiatric hospitalization. Its applicability to non-violent offenders or general psychiatric populations is limited.
Strengths & limitations
- Actuarial methodology produces transparent, empirically derived probability estimates; scores are replicable across raters and less subject to clinical bias than unstructured judgment.
- Substantial research support: the VRAG has been cross-validated in multiple independent samples of offenders, with demonstrated predictive validity and area under the curve (AUC) of 0.71–0.73 in most studies.
- Incorporates both static historical factors (psychopathy, violence history, marital status) and dynamic elements, offering a balanced risk profile.
- Brief administration (30–60 minutes) and file-based scoring make it feasible in high-volume forensic settings.
- Widely recognized in North American criminal justice and forensic psychiatric systems; results are generally acceptable in legal proceedings.
- Derived exclusively from adult male offenders in Canadian forensic psychiatric settings; generalizability to females, younger offenders, or non-psychiatric criminal populations is poor and not empirically established.
- Does not account for recent changes in mental health treatment, medication response, or environmental factors; scores remain stable despite material clinical improvements.
- Produces group probabilities, not individual predictions; misinterpretation of bin-level recidivism as individual risk is common and legally risky.
- Dependent on complete and accurate historical records; missing or inaccurate data on prior violence or offense history compromises scoring validity.
- Limited applicability in non-English-speaking or non-North American forensic systems; cross-cultural and international validation is sparse.
- Does not incorporate protective factors (employment, family support, treatment engagement); balanced risk assessment may require concurrent use of protective factor measures (e.g., SAPROF).
Frequently asked
What does a VRAG score of +15 mean? Does it predict that this person will reoffend?
A score of +15 places an individual approximately in bin 6, associated with a group recidivism rate of about 45–50% over 5 years. This is a group-level statistic, not an individual prediction. It means individuals with similar scores and characteristics exhibited ~45–50% recidivism in the derivation sample, but we cannot say this specific person has a 45–50% chance. Individual risk depends on many factors not captured by the VRAG (e.g., current employment, family relationships, recent treatment response).
Can the VRAG be used to assess risk for violent behavior in the community (not just recidivism)?
The VRAG was specifically developed to predict violent recidivism (reoffense) among individuals released from forensic psychiatric institutions. While violent recidivism is a proxy for community violence risk, the VRAG is not designed to assess imminent danger or acute violence risk in general psychiatric settings. For acute risk assessment, structured professional judgment tools (e.g., HCR-20v3) are more appropriate.
Should the VRAG be updated periodically?
The original VRAG score does not change—it reflects historical items (age at offense, prior violence history, psychosis diagnosis) that are fixed. However, new information may warrant reassessment using an updated version or recomputation if base factors (e.g., marital status, substance use history) are discovered to be inaccurate. Dynamic clinical factors (current symptoms, employment) are not part of the VRAG scoring but should inform clinical judgment and risk management decisions alongside the VRAG score.
Can women or youth be assessed with the VRAG?
The VRAG was developed and validated exclusively in adult males and should not be used to assess females or youth. Predictive validity is not established in these populations, and research suggests violence risk factors differ substantially by gender and age. Sex-specific instruments (e.g., for female offenders) or age-appropriate tools (e.g., VORAG for adolescents) should be used instead.
Is the VRAG still used, or has it been replaced by newer tools?
The VRAG remains in use in many North American correctional and forensic psychiatric systems, particularly in established institutional protocols. However, newer structured professional judgment tools (e.g., HCR-20v3) have gained favor in clinical practice because they incorporate dynamic factors and protective elements more explicitly. Many forensic organizations now use VRAG alongside or supplementary to SPJ instruments for a more comprehensive assessment.
Sources
- Harris, G. T., Rice, M. E., & Quinsey, V. L. (1993). Violent recidivism of mentally disordered offenders: The development of a statistical prediction instrument. Neuropsychiatry Neuropsychology and Behavioral Neurology, 6(4), 269–279. DOI: 10.1177/0093854893020004001 ↗
- Harris, G. T., Rice, M. E., Camilleri, J. A., & Boer, D. P. (2015). Prospective, multi-site study of adolescent violent recidivism: The Violent Offense Risk Assessment Guide. Canadian Journal of Criminology and Criminal Justice, 57(4), 486–524. link ↗
How to cite this page
ScholarGate. (2026, June 3). Violence Risk Appraisal Guide. ScholarGate. https://scholargate.app/en/forensic-psychology/violence-risk-appraisal-guide
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- HCR-20v3Forensic Psychology↔ compare
- LSI-RForensic Psychology↔ compare
- SAPROFForensic Psychology↔ compare
- Static-99RForensic Psychology↔ compare