Background

Violence is an increasingly common and significant problem for youth worldwide. Youth who rely on treatment at urban EDs are more likely to die as the result of violence than any other disease/condition for which they seek care. The first step in helping youth that are at risk, is identifying them. We developed a 7–item screening tool called VPET. The purpose of this study is to validate the VPET screening tool in identifying high-risk youth.

Methods and findings

We prospectively enrolled a convenience sample of children during the index ED visit who were called 3 months and 6 months after this visit. 269 youth (33%) completed 3-month follow up (44.2% male); 240 youth (29.4%) completed 6-month follow up (45% male); 84.0% reported some level of violence exposure after 3-months and 84.2% (n = 240) reported some level of violence exposure after 6-months.

Predictive validity was assessed by Spearman's correlation between VPET score and follow-up score at 3-months and 6-months post-enrollment; logistic regression to calculate odds ratios between positive VPET item responses and positive follow-up score; ROC curve analysis. VPET score had internal consistency, as tested by Cronbach's alpha (α = 0.642). Children who were male, non-white, and had been hurt at home or school reported the highest VPET scores.

Conclusion

VPET has sufficiently strong psychometric function and performs well as a screening tool to predict future violence exposure for youth ages 8–17. Five questions on the VPET screening tool are independently predictive of violence reported at 6 months and four questions at 3 months.