Assessing Risk for Future Firearms Violence in Young People Who Present to ED
EXECUTIVE SUMMARY
A new clinical index tool designed specifically for the emergency environment predicts the risk for future firearms violence in young people 14-24 years of age. The approach employs a brief, 10-point instrument that can be administered in one to two minutes, according to investigators. They also note that while the tool is based on data from a single ED in Flint, MI, the tool should be applicable to urban EDs in regions that have similar characteristics.
- To create the tool, investigators used data from the Flint Youth Injury Study, an investigation of a group of patients 14-24 years of age who reported using drugs in the previous six months and accessed care at a Level I trauma center.
- Using a machine learning classification approach, investigators combed through the data, finding that the most predictive factors for firearm violence could be categorized into four domains: peer and partner violence victimization, community violence exposure, peer/family influences, and fighting.
- Ideally, investigators note the tool will be employed along with interventions targeted toward patients at high risk for future firearms violence.
The CDC reports that people 15-24 years of age are particularly vulnerable to firearm violence. Not only is homicide the third leading cause of death in this group, but most of these deaths stem from gunfire.
Further, emergency providers in urban areas know well the burden of caring for young patients who present with non-fatal firearm injuries. Medical and lost-work costs related to victims who presented to EDs with these injuries in 2010 neared $3 billion, according to government data.
Although prevention services potentially can short-circuit the cycle of violence often seen in young people, it is not possible to provide such help to every at-risk individual who presents to the ED for care. However, researchers at the University of Michigan (UM) School of Medicine in Ann Arbor, have developed a new clinical risk tool that could make this job easier.1 Designed specifically for the emergency care setting, researchers say the tool is very brief, enabling providers to quickly identify the young emergency patients who are most vulnerable to future firearm violence so preventive interventions can be applied where they are needed most.
Tool Focuses on Firearm Violence
Investigators focused specifically on developing a tool for the ED because this setting provides both a key access point for identifying youth at risk for firearm violence and the opportunity for a “potential teachable moment,” according to Jason Goldstick, PhD, the lead author of the study and research assistant professor in the department of emergency medicine at UM School of Medicine.
Researchers report that they looked at existing screening tools that assess for risk of violence, but noted that the instruments lacked a focus on firearms and took too long to administer in the emergency setting.
To create the new tool, investigators used data from the Flint Youth Injury Study,2 an investigation of a group of patients 14-24 years of age who reported using drugs in the previous six months, and accessed care at a Level I trauma center. That study, which compared patients who were victims of assault with patients seen in the ED for other reasons, included 599 participants who completed a 115-item survey that contained questions pertaining to violence, alcohol and drug use, and the influence of peers.
“The study participants were assessed five times over two years. Using that data, we were able to determine which of the participants [experienced] gun violence, either as a victim or perpetrator, during that two-year period,” Goldstick explains. “We then used measurements taken at the start of the study to see which [factors] were most predictive of [future] firearm violence.”
Using a machine learning classification approach, investigators found that the most predictive factors could be categorized into four domains: peer and partner violence victimization, community violence exposure, peer/family influences, and fighting. Consequently, the researchers took one item from each domain to create a 10-point score that is specific to firearm violence. They named the score SaFETy, so that it could be used as a mnemonic to help providers remember the four items:
- serious fighting frequency (0-4 points, depending on frequency in the previous six months);
- friends who carry weapons (0-1 point, depending on the number);
- community environment (0-1 point, depending on the frequency of hearing gunshots in the previous six months);
- firearm threats (0-4 points, depending on how frequently someone was threatened with a gun).
In the study, every increase of a single point in the SaFETy score was indicative of an increased risk of firearm violence, with a score of 8 or more being 100% specific for predicting risk for future firearm violence.
External Validation Needed
Goldstick acknowledges that emergency providers are loathe to take on any additional screening responsibilities, but notes that this tool is very specific to the needs of the ED, both because of the study population used to devise it and the fact that it was designed to be very brief.
“We believe this score could be administered in one to two minutes,” he says.
Goldstick adds that two of the key collaborators on the study are emergency physicians, including the principal investigator, Rebecca Cunningham, MD, a professor of emergency medicine at UM, and the lead author of the Flint Youth Injury Study.
“That is the primary reason why we placed such great priority on making this risk assessment tool as brief as possible,” he says.
However, Goldstick allows that the tool might not be applicable to every patient population.
“This was a tremendously high-risk study population. Over half of [the study participants] we could ascertain had firearm violence, as a perpetrator or a victim, during the two-year follow-up period. Therefore, we have no evidence that this index would be useful in, for example, a suburban emergency department,” he observes. “[However], while additional validation is required, there is reason to think it would be applicable for risk stratification at EDs in urban centers similar to Flint, MI, such as Youngstown, OH, Camden, NJ, and Oakland, CA, when applied to similarly aged individuals.”
At press time, the SaFETy score had just been unveiled, and had not yet been administered beyond the scope of the study, but investigators look forward to opportunities to validate the score externally, and to determine how it can be modified to suit different populations, Goldstick explains.
“More broadly, we envision coupling this type of risk stratification with appropriate levels of preventive services which ... means that such services need to be developed,” he says. “Effectively utilizing those resources requires a basis for risk stratification, and that is the piece of the puzzle this work is meant to provide.”
REFERENCES
- Goldstick JE, Carter PM, Walton MA, et al. Development of the SaFETy Score: A clinical screening tool for predicting future firearm violence risk. Ann Intern Med 2017; doi: 10.7326/M16-1927.
- Cunningham RM, Carter PM, Ranney M, et al. Violent reinjury and mortality among youth seeking emergency department care for assault-related injury: A 2-year prospective cohort study. JAMA Pediatr 2015;169:63-70.
SOURCE
- Jason Goldstick, PhD, Research Assistant Professor, Department of Emergency Medicine, University of Michigan School of Medicine, Ann Arbor, MI. Email: [email protected].
A new 10-point scale offers potential intervention methods for patients most in need of prevention services.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.