Abstract

BACKGROUND

Although there has been a downward trend in crime over the past two decades, violence-related injuries are a growing concern for many urban areas in the U.S. The number of homicides in the 50 largest U.S. cities increased by 17% from 2014 to 2015 and nearly 1.5 million patients were treated for nonfatal assaults in 2014 at U.S. emergency departments.¹ In addition to human costs, such as health disparities and excess mortality, violent injury also results in high financial costs to society. The average cost of medical care for treating a patient hospitalized for violent injury has been estimated to be approximately $25,000.² When accounting for indirect costs of injury, such as lost productivity and disability, the price is significantly greater with national estimates exceeding $30 billion annually.² Finally, in addition to physical wounds, the effects of violent injury can result in long-term mental and physical health conditions and a decrease in quality of life.³

Hospital-based violence intervention programs (HVIP) aim to reduce both retaliatory injury and recidivism by providing intensive, case management services to high risk patients who were violently injured.⁴ Violently injured patients often face numerous obstacles after being discharged from the hospital including: accessing follow-up care, finding safe and adequate housing, returning to work or school, addressing legal issues, and managing posttraumatic stress and community pressure to retaliate.⁵ These obstacles can lead patients to continue to engage in behaviors that increase risk for re-injury, such as substance use, weapon carrying, or illegal activities.⁶ Violent injury recidivism has been estimated to be as high as 44% in the 5 years subsequent to an assault resulting in hospitalization.⁷ Young adults who are seriously injured in an assault are nearly twice as likely to have another violent injury requiring hospital treatment within two years compared to their counterparts with non-violent injuries.⁸ HVIP operate from the premise that there is a unique window of opportunity to effectively engage with victims of violent injury while they are recovering in the hospital.⁴ These programs often provide a broad range of services including medical, psychological, legal, and financial counseling in order to reduce criminal involvement and re-injury.^5,9–12

Although published studies demonstrate that HVIP are effective, to date, few studies have been undertaken to identity which components of HVIP contribute to recidivism reduction and how long their effects last.⁵ Additionally, prior studies have been limited by the fact that HVIP typically rely on self-report or data within their own healthcare system to identify new injuries. Our study aims to improve upon previous work by measuring recidivism over an 8 year period and identifying injuries that are treated outside the HVIP-affiliated trauma center. Using data from the one of largest regional health information exchanges in the world, our study addresses limitations of prior studies by evaluating data that contains emergency department encounters for nearly the entire state.^13,14 Additionally, using this resource we are able assess participants’ recidivism since the inception of the HVIP program, Prescription for Hope (RxH), over 8 years ago.

The objective of this study was to quantify the long-term recidivism rate of participants in an HVIP program using data from a regional health exchange. We hypothesized that some HVIP participants would be treated for new violent injuries outside of the original trauma center they were initially treated at. We chose to examine this because patients in urban areas have many providers that can treat their injuries and the provider they receive care from could depend on the location they are injured, how severe the injury is, and who transports them. The results of this study can provide evidence for the effectiveness of HVIP over time and may offer insight into other medical needs violently-injured patients have after trauma center discharge.

METHODS

RESULTS

Between January 2009 and August 2016, The RxH program enrolled 328 patients. We were able to identify 317 patients (96.6%) in the INPC database. Of these 317 patients, 242 patients (76.3%) had ED visits. Patients had 1,575 unique encounters at 11 different EDs. Based on diagnosis codes and the patient’s stated reason for visit, the most common reason for being treated at the emergency department was due to pain (718 encounters). The second most common reason was due to suspected complications or the need for additional postoperative care (181 encounters). Encounters were also commonly coded for substance use (110 encounters for alcohol abuse, 76 for tobacco use, and 24 for other types of drug use). We also identified 79 encounters related to chronic medical conditions (e.g., diabetes, hypertension, asthma, etc.), 42 for chronic pain, and 14 encounters for suicidal ideation. Encounters for accidental injuries were also common in this cohort, with 37 visits related to motor vehicle crashes and 55 related to falls. (Figure 1)

Open in a separate window

Figure 1

ED Encounter Types and Frequencies

Pain was the most common reason for visiting the ED, followed by other injury-related complaints, and suspected surgical complications or additional follow-up care needed. Other co-morbid conditions were also common. “Injury-related” visits were a general category that captured conditions such as fracture, abrasions, contusions, etc. If a mechanism of injury was specified, such as a fall, then the encounter was categorized as such. Because encounters often have multiple diagnoses, these categories are not mutually exclusive. For example, a single encounter may be coded as both injury-related and alcohol abuse, if both were indicated in the ED data.

We flagged 288 encounters for additional review based on diagnosis codes and the listed reason for ED visit. There were 1,949 provider notes associated with the flagged encounters in the INPC. There were 40 encounters with no notes available, 93 with notes that did not specify what happened to the patient, and 4 notes where we could not ascertain whether an old or new violent injury was being treated. We identified 17 encounters for new violence-related injuries, which were treated at 3 different hospitals, all belonging to different health systems. Nine of these injuries were treated at the hospital associated with the RxH program, whereas 8 were treated at other institutions. A total of 15 patients recidivated (4.4%), with two participants sustaining multiple violence-related injuries during the follow-up period. Of these 17 encounters, 5 were admitted to inpatient care and 12 were discharged from the ED. Most encounters were the result of physical assault (10), however, 6 new injuries were caused by firearms and 1 was due to stabbing. None of the encounters indicated that the patient died, however, based on an obituary search, we are aware of at least one participant who is deceased.

There were no significant differences in regards to age, gender, race/ethnicity, mechanism of injury, education level, employment at time of injury, and program goal completion between those who recidivated and those who did not. (Table 2)

DISCUSSION

Overall our study demonstrated that the recidivism rate for RxH participants remains low compared to published U.S. data.⁷ Our results show that over half of new violence-related injuries were treated outside of the HVIP-affiliated trauma center, indicating that relying on single-institution data to evaluate HVIP programs may be unreliable. We also found that most new injuries were relatively minor compared to the index injury and were treated and released from the emergency department.¹⁴ In the first year of the RxH program, we reported a 2.9% 1-year recidivism rate.¹⁵ The current study found a 4.4% recidivism rate based on 8 years of statewide data, which indicates that RxH has an enduring positive effect on the vast majority of participants.

Our study also demonstrates that violently-injured patients have a variety of medical issues that are being treated in emergency departments. Pain was the most common reason for ED visits which may indicate patients are not being discharged with adequate pain medication or that they are developing chronic pain from their injuries. It is also possible that they are either misusing pain medication or selling it as a means to provide income. Finally, our data shows that patients often receive follow-up care for their injuries in an ED setting, suggesting there is a need to improve access to outpatient care after hospital discharge. This may indicate that discharge instructions regarding wound care may be unclear or symptoms of infection are not well understood by patients and caregivers. Future studies on what prompts these types of ED encounters are needed and this information could be applied to improving the discharge planning process.

Our study shows that HVIP patients are also susceptible to other types of injuries, including overdoses, suicides, motor vehicle accidents, and falls. This suggests the need to broaden the focus beyond violence-related injuries in this population. In particular, we found that patients who attempt suicide often do so beyond the typical 1 year HVIP follow-up period. Additionally, suicidal patients tend to have more ED encounters and present with other behavioral health issues.

Conclusion

Based on our results and those of other studies, we conclude that HVIP, and the RxH program in particular, are effective at decreasing recidivism of violent injuries and can sustain their effects over many years. However, adequate evaluation of HVIP likely requires access to data from multiple institutions that contain detailed information on encounters in the form of provider notes. Furthermore, HVIP may benefit patients by seeking out partnerships with organizations that work to prevent suicide, substance use disorders, and other unintentional injuries.

Acknowledgments

Footnotes

References