Intentional firearm-induced injuries have increased as a cause of death among adolescents. Because screening for other high-risk behaviors such as drug and alcohol abuse, domestic violence, and failure to use seat belts has improved patient outcomes, screening for carrying firearms also may be useful. Direct questioning of a patient about high-risk behavior is less useful than formalized screening tests, because sensitivity is decreased and risk stratification is difficult to perform in the former. There are many known risk factors for carrying firearms, including male sex, smoking, drug use, alcohol use, poor academic performance, being older than classmates, television viewing, having multiple sexual partners, feeling threatened, fear, and poor family relations. These risk factors potentially could be used to develop a screening tool for firearm carrying.

Because screening would increase the likelihood of counseling for high-risk patients, perhaps suicides resulting from firearm carrying could be reduced. Hayes and Sege propose a screening tool developed from the National Youth Risk Behavior Survey, a cross-sectional surveillance survey performed by the Centers for Disease Control and Prevention at schools in 52 representative communities across the country.

The National Youth Risk Behavior Survey notes five risk factors that correlate with firearm carrying. These risk factors were used to develop a brief FiGHTS (Fighting, Gender, Hurt while fighting, Threatened, Smoker) score. An extended FiGHTS score also was developed that includes items on substance abuse, and sexual and criminal behavior. These risk characteristics were validated by large variations of the National Youth Risk Behavior Survey data sets.

Scoring the questionnaire involved giving one point for each question answered affirmatively. The group of adolescents with FiGHT scores of 2 or higher included a majority (82 percent) of the gun carriers. The positive predictive value of the screening tool is low to moderate because of the low incidence of gun carrying among adolescents, but the negative predictive value of a low score is higher than 90 percent. The extended FiGHT score also had good positive predictive value for recent firearm carrying risk and appears to predict absolute risk.

The authors conclude that the FiGHTS screening tool offers a generalizable way to stratify adolescent risk of firearm carrying, especially by identifying those with low risk. The brief FiGHTS score is easy to administer and provides coarse risk stratification. The extended FiGHTS score is useful as a predictor of absolute probability of recent firearms carrying.

In an accompanying editorial, Larkin notes important risk factors for firearm carrying that are missing from the screening tool, including intravenous drug use and television watching. He notes the advantages of the extended FiGHTS tool but recognizes the need to have a streamlined screening process that might use computerized screening with provisions for response-dependent information about help options. Decreasing adolescent firearm violence is a major objective of Healthy People 2010, making further work in this area imperative.