There is a growing recognition that emergency departments (ED's), especially those in large urban hospitals, can provide an effective setting for violence risk factor detection and intervention (Ketterlinus, 2008). In addition to the treatment of physical injuries in the intentionally injured youth, the ED offers a potential opportunity to explore psychosocial needs and to provide early intervention, with the aim of preventing future violence in the form of reinjury or retaliation. The significant life disruption and physical suffering brought about by a violent injury provide a unique teachable moment during which a youth may feel vulnerable and more receptive to a message of change (Cunningham, et al., 2005).
Youth violence has been shown to be a chronic, recurrent disease, with victims being at increased risk for repeat injury and retaliation. After their physical injuries have been treated, youth violence victims are oftentimes discharged from the hospital to the same high-risk environments in which they were victimized. Reported hospital readmission rates of violently injured youth have ranged from 6-44% (Snider & Lee, 2009). In one study (Sims, et al., 1989), urban violence victims admitted to a hospital trauma service had a 20% rate of becoming victims of homicide within 5 years. Youth victims of assault not only are likely to be revictimized, but are also significantly more likely to become perpetrators of future violence (Rivara, et al., 2005).
Beginning in the early- to mid-1990's, a number of hospital- and ED-based youth violence intervention programs began to take root in U.S. urban settings. Such programs have now been implemented in at least 8 major cities. While these programs vary in their structure as well as the nature, intensity and duration of provided services, they share in common the basic approach of providing psychosocial evaluation to youth violence victims while in the hospital, assessing safety and retaliation risks, and referral to appropriate community-based resources. Such community resources include individual and family counseling, support groups, life skills training (e.g., anger management, conflict resolution), mentoring programs, academic support, chemical dependency programs, mental health treatment, employment training and legal support (Worden & McLean, 2008).
One of the earliest hospital-based programs is the Caught in the Crossfire Program in Oakland, California, implemented in 2004, and now considered a model violence prevention program. Becker, et al. (2004) demonstrated that six months after their injury, the youth enrolled in the intervention arm were 70% less likely to be arrested for any offense and 60% less likely to have any criminal involvement than the control group. However, the authors found no statistically significant differences for rates of re-injury or death. My major critique of this study is that it did not perform random assignment of hospitalized youth, but instead compared youth from two different time periods, allowing for potential confounding of the results. In addition, the subjects were only followed for six months, leaving in question the longer-term outcomes of the intervention.
Twelve months after enrollment in a Chicago hospital's youth violence intervention program, the treatment group self-reported a significantly lower incidence of repeat violence (8.1%), compared with the control group (20.3%) (Zun, et al., 2006). There were, however, no significant differences in rates of return to the ED, arrest, or incarceration. While this study did randomize its patients, the nature of the intervention precluded blinding of the assignment process.
Marcelle and Melzer-Lange (2001) reported that, of more than 200 patients aged 10-18 who had been enrolled in Project Ujima, a Milwaukee ED-based violence intervention program, only 3 (1%) returned to the ED as a result of violent injury one year after enrollment. However, the authors did not follow non-enrolled youth, precluding comparison with a control group.
In a randomized control trial in a Baltimore pediatric ED, Cheng, et al. (2008) found that those youth enrolled in a violence intervention program had no significant difference in rate of repeat injury or community service utilization after enrollment in a violence intervention program, compared to controls. However, the study was limited by large loss to follow-up, and by the fact that the intervention often did not begin until weeks after the injury, potentially missing a short-lived teachable moment opportunity immediately after injury.
Cooper, et al. (2006) found that those youth enrolled in a Baltimore hospital violence intervention program had a significantly lower rate of hospital recidivism for repeat violent injury (5%) than the control group (36%). Youths from the treatment group also had a higher rate of employment (82% vs. 20%), were one-fourth as likely to be convicted of a violent crime, and had significantly shorter times of incarceration. The authors also made a strong case for cost effectiveness of the intervention. They estimated the aggregate cost of hospitalization to be $138,000 for the 56 youth in the treatment group versus $736,000 for the 44 youth in the treatment group. They further predicted that the total cost of incarceration would be $450,000 for the treatment group versus $1,700,00 for the control group. A strength of this study is its randomization process. However, as in the Zun (2006) study discussed above, the nature of the intervention prevents blinding of the study.
A limitation, common to all studies cited above is large loss to follow up of enrolled patients. Many of these youth have transient and shifting social situations that make extended follow up challenging. In addition, these studies tend to have relatively low numbers of subjects and limited follow up periods, highlighting the need for ongoing research. As the study of hospital-based interventions evolves, it will also be informative to elucidate the specific intervention features of individual hospital programs that prove to be the most effective in preventing re-injury and future criminal involvement.