Predictive validity of the HCR-20 for inpatient self-harm

Area of Science:

• Forensic Psychiatry
• Clinical Psychology
• Risk Assessment

Background:

• Limited instruments exist to assess self-harm risk in psychiatric patients.
• Evidence for the predictive validity of self-harm risk assessment tools is scarce.
• Self-harm can precede other-directed aggression and violence.

Purpose of the Study:

• To evaluate the predictive validity of the HCR-20 violence risk assessment for self-harm.
• To identify patient groups for whom the HCR-20 is a more effective predictor of self-harm.

Main Methods:

• A pseudo-prospective cohort study involving 504 patients in a UK secure/forensic mental health setting.
• Routinely collected data, including HCR-20 assessments and self-harm incidents within 3 months post-assessment.
• Statistical analysis to determine the predictive power of HCR-20 scores and subscales for self-harm.

Main Results:

• The HCR-20 total score, H10 and R5 subscales, and SJ for violence significantly predicted self-harm.
• Area Under the Curve (AUC) values ranged from .345 to .749, indicating moderate predictive effect sizes.
• Personality disorder and impulsivity were strong predictors; the R5 scale had the most relevant items.
• Predictive efficacy was greater for women, and patients with schizophrenia or personality disorder.

Conclusions:

• The HCR-20 demonstrates significant predictive value for self-harm in psychiatric settings.
• Supplementing HCR-20 ratings with clinical judgment and known self-harm risk factors can enhance risk assessment.
• This approach may reduce the need for multiple, specialized assessment tools for self-harm.