Re-Examining the Predictive Validity and Establishing Risk Levels for the Dynamic Appraisal of Situational...

Area of Science:

• Psychiatry and Mental Health
• Nursing Practice
• Risk Assessment Tools

Background:

• The Dynamic Appraisal of Situational Aggression: Youth Version (DASA:YV) is a validated tool for assessing imminent violence risk in youth.
• Current recommendations suggest linear interpretation of DASA:YV scores for intervention.
• Re-analysis of existing data is needed to refine risk categorization and predictive validity.

Purpose of the Study:

• To re-examine the predictive validity of the DASA:YV using contemporary data analytic methods.
• To determine and validate appropriate risk bands for DASA:YV scores.
• To enhance the practical utility of the DASA:YV for nursing staff in youth settings.

Main Methods:

• Utilized mixed-effect logistic regression models to analyze correlated observational data.
• Employed two approaches for identifying risk bands: adapting adult DASA categories and using Chi-square automated interaction detection (CHAID).
• Compared the predictive accuracy (AUC) of three- and four-category risk band models.

Main Results:

• A three-category risk band model demonstrated comparable predictive validity (AUC = 0.85) to a four-category model.
• No statistically significant difference was found between three- and four-category risk band models.
• The study supports the recommendation of a three-category approach for DASA:YV interpretation.

Conclusions:

• A three-category risk band system is recommended for the DASA:YV, offering more accurate risk categorization.
• Implementing these refined risk bands can aid nursing staff in timely and appropriate interventions.
• Early identification of escalating risk may reduce the need for restrictive practices in youth mental health settings.