Analysis of intentional and unintentional injuries caused by firearms and cutting/piercing instruments among Providence youth, Nov 2004-Dec 2007.

Despite declining rates of violent deaths over the last decade, homicide remains among the top 4 causes of death for Americans younger than 34 years old. Morbidity from youth violence is also significant. In 2004, injuries from youth violence resulted in over 750,000 visits to emergency departments (EDs) nationwide. In Rhode Island, 28% of RI high-school students reported at least one physical fight in the last year, and according to hospital discharge data over 100 RI adolescents are hospitalized each year due to injury from assault. Risk factors for commission of youth violence are well-established; e.g., male gender, low socioeconomic status, availability of weapons, use of alcohol or drugs, and age between 12 and 20. 5-7 Demographic and behavioral correlates of injury from youth violence are not as clearly determined. Studies establish only that the best predictor of youth injury from violence is a previous violent injury. Alcohol use and other known adult risk factors for violent injury do not necessarily apply to youth. 16 Moreover, studies of youth injury predictors rely on inpatient samples, and may be subject to bias from selection of a more severely injured population. Studies which rely on hospital discharge data may be subject to coding inaccuracies; some research suggests that erroneous external cause of injury codes are often given to youth injuries. 18 EDs, in contrast, represent a wider spectrum of injury severity and type. This study was designed to address the lack of published data on trends of youth violent injury in Providence, as well as the need for additional data about risk factors and correlates of violent injury. We examined the distribution of intentional and unintentional injury from weapons (cutting/piercing and firearm) among Providence youth, the correlates and predictors of these violent injuries, and the accuracy of our ED coding of violent injury. METHODS In this retrospective cross-sectional chart review of cutting/piercing and firearm injuries among Providence youth (younger than 21 years old), we examined all appropriate youth injuries presenting to Hasbro and Rhode Island Hospital EDs between November 1, 2004, and December 31, 2007. These hospitals provide the only adult and pediatric Level I trauma services in the state; they also have the highest ED volumes in the City of Providence. The Rhode Island Hospital/ Lifespan Institutional Review Board approved the protocol. Within the study time period, cases were selected from the ED billing databases using ICD-9-CM external cause of injury codes (E-codes): all cases corresponding to injuries from sharp (cutting/piercing) objects (E920.0-.9, E956, E966, E986, E974) or from firearms (E922.0-.9, E922.8. E922.9, E955.0-.4, E965.0.5, E968.3, E979.4, E985.0.6, E970) were retained. Ecodes corresponding to assault with unknown object (E968.9) were also included, in case these assaults had occurred with a weapon but were miscoded. These codes encompass both unintentional and intentional injury. The first valid, relevant Ecode was selected as per national guidelines for injury surveillance. Records were further selected by age (< 21) and zip-code (02903, 02904, 02905, 02906, 02907, 02908, 02909, 02912, and 02940). Through this process, 446 charts were identified. Two researchers, using a standardized chart review form and protocol, reviewed each chart. In the case of discrepancies or questions, the reviewers consulted with each other and, when necessary, with a third member of the research team to reach a consensus about coding. All data were entered into a password-protected Excel database.