CLINICIAN'S CAPSULE
What is known about the topic?
Trauma recidivism rates range from 1% to 44%, yet the rate of major trauma recidivism is unknown at a provincial level.
What did this study ask?
What is the rate of major trauma recidivism in Nova Scotia, and what factors are associated with recidivism and mortality?
What did this study find?
Major trauma recidivism was infrequent; however, the adjusted odds of mortality were over three times greater for recidivists.
Why does this study matter to clinicians?
Screening and brief intervention in cases of major trauma involving violence or substance abuse may help reduce rates of recidivism.
INTRODUCTION
Major trauma recidivists are patients with a prior severe traumatic injury.Reference Reiner, Pastena, Swan, Lindenthal and Tischler1 Previous investigations of recidivism observed rates from 1%–2%Reference Kaufmann, Branas and Brawley2, Reference Allan, Davis and Pandya3 to 44%.Reference Sims, Bivins and Obeid4 In the only other Canadian study on recidivism to date, Caufeild and colleagues found an overall rate of 0.38% at two trauma centres in Toronto.Reference Caufeild, Singhal and Moulton5
Recidivists have higher long-term mortality after discharge,Reference Strong, Greene and Smith6 and the mortality risk increases with each successive admission.Reference Brooke, Efron, Chang, Haut and Cornwell7 Surprisingly, some have observed better hospital outcomes among recidivists.Reference Dixon, Como, Banerjee and Claridge8 Recidivism is associated with a younger age, male gender, a penetrating injury, violence, and substance abuse.Reference Kaufmann, Branas and Brawley2, Reference Caufeild, Singhal and Moulton5, Reference Strong, Greene and Smith6 Our objective was to determine the major trauma recidivism rate in Nova Scotia. We compared outcomes between recidivists and non-recidivists and assessed for factors associated with recidivism and mortality.
METHODS
We reviewed all adult patients (>17 years) who sustained a major trauma using data from the Nova Scotia Trauma Registry, a provincial population-based trauma registry under the Nova Scotia Department of Health and Wellness. The criteria for registry inclusion were: an Injury Severity Score of ≥12, an appropriate International Classification of Diseases (ICD) External Cause of Injury Code, penetrating injuries with an Injury Severity Score of ≥9, all trauma team activations, and deaths (at the scene, in the emergency department, and within 24 hours of admission). The staff of Trauma Nova Scotia identified patients meeting the registry inclusion criteria by reviewing records at all regional hospitals and tertiary care centres. Scene and pre-hospital deaths were identified by reviewing records from the Nova Scotia Medical Examiner Service. Trauma Nova Scotia considers a blood alcohol concentration of ≥2 mmol/L to be positive, as defined in the National Trauma Registry (NTR) data dictionary.
Major traumas between April 1, 2001, and March 31, 2015, were eligible. Individuals with more than one major trauma during the study period were considered recidivists. The primary outcome was the recidivism rate. Secondary outcomes were the length of stay (in-hospital, intensive care unit (ICU), and intermediate care unit), the occurrence of complications (any ICD-10-CA or NTR complication arising after hospital observation and influencing patient management or treatment), and mortality (including the scene, pre-hospital, and in-hospital deaths). Recidivists and non-recidivists were compared using Student's t-tests and a chi square analysis. We created two multivariate logistic regression models to assess for factors associated with recidivism and mortality. For recidivists, variables were based on their most recent injury. All variables were entered in a single step as a block. Adjusted odd ratios (ORs) are reported for measurement of the effect size with 95% confidence intervals (CIs). All analyses were performed using IBM SPSS Statistics (version 23; IBM Corp; Armonk, NY).9 Ethical approval was obtained from the Nova Scotia Health Authority Research Ethics Board.
RESULTS
Of 9,635 adult patients who sustained a major trauma, the recidivism rate was 2% (150/9635). Six patients had more than two traumas; no patient had more than four traumas. The mean duration between the initial and second injury was 41 ± 40 months (range 1–159 months). Patient characteristics and outcomes are shown in Table 1. Recidivists were more likely to be male and have a lower Injury Severity Score on emergency department (ED) arrival. Falls, motor vehicle collisions, and gunshots/violence were the most common injury causes (Supplementary Figure 1). More non-recidivists were injured in motor vehicle collisions, as compared with the index injury of the recidivists (25% v. 18%, respectively; p = 0.047), and more recidivists were injured by gunshots/violence (index injury), as compared with non-recidivists (15% v. 9%, respectively; p = 0.014). Among recidivists, alcohol testing was performed in 43% (65/150) at the initial injury, with 58% (38/65) being alcohol positive. Alcohol testing rates increased over the study period, especially in patients with a trauma team activation (Supplementary Figure 2). This increase can be attributed to strong leadership by the provincial trauma program, forcing a shift in culture toward routine testing for alcohol.
Table 1. Characteristics and outcomes of recidivists and non-recidivists, 2001 to 2015
AIS = Abbreviated Injury Scale; BAC = blood alcohol concentration; ICU = Intensive Care Unit; ISS = Injury Severity Score; IMCU = Intermediate Care Unit; LOS = length of stay; n/a = not applicable; SD = standard deviation; TBI = traumatic brain injury; TTA = Trauma Team Activation.
*Missing values: Age (n = 157), ISS (n = 164), AIS Head score (n = 4391)
†T-test or chi square analysis comparing non-recidivists with initial or recurrent injury of recidivists.
Mortality was 31% for both recidivists and non-recidivists. Recidivists had a longer in-hospital length of stay on recurrent injury (26 ± 48 days v. 16 ± 35 days, respectively). There were no differences in the length of stay in the ICU or intermediate care unit. Multivariate logistic regression models were created to assess factors associated with trauma recidivism and mortality (Supplementary Table 1). After adjusting for potential confounding factors, the likelihood of mortality was over three times greater for recidivists, as compared with non-recidivists (odds ratio [OR] 3.67, 95% CI 2.06–6.54). Recidivism was more likely in younger patients (OR 0.98, 95% CI 0.97–1.00) and patients for whom the trauma team was activated (OR 0.59, 95% CI 0.34–0.96).
DISCUSSION
This is the first Canadian study of major trauma recidivism on a provincial level. The 2% rate in Nova Scotia is greater than the 0.38% rate in Toronto.Reference Caufeild, Singhal and Moulton5 We collected data from a provincial trauma registry; Caufeild et al. used registry data from two trauma referral centres (neither captured deaths at the scene or pre-hospital). If traumas of lesser severity are included, recidivism rates are considerably higher; a study from Sweden reported a 36% rate of recurrent moderate/severe injuries.Reference Röding, Lindkvist, Bergström, Svensson and Lysholm10
In addition to the negative effects of recidivism on the patient and their loved ones, recidivism is a significant burden on the health care system.Reference Hayman and Crandall11, Reference Maier12 Hospitals and trauma centres are often the initial point of contact for patients dealing with violence or substance abuse; screening and brief intervention for these patients can reduce injury recurrence.Reference Gentilello, Rivara and Donovan13, Reference Aboutanos, Jordan and Cohen14 A brief intervention for alcohol and illicit drug use reduced the incidence of trauma recidivism by 52%.Reference Cordovilla-Guardia, Fernández-Mondéjar and Vilar-López15 In our study, alcohol screening rates varied greatly, and a toxicology screening was performed infrequently; thus, our findings related to substance use must be interpreted with caution. With Canada set to legalize marijuana in 2018, future investigations are planned to determine the impact of legalization on trauma recidivism. There may be a need to revisit and expand mandatory testing in the ED to include cannabis and alcohol in all trauma team activations.
This study was subject to inherent limitations of research on an existing dataset. Although data were collected from a robust provincial trauma registry, some variables had missing values. The registry does not capture socioeconomic status, ethnicity, history of criminal activity, or psychiatric illness. Furthermore, the registry is limited to major traumas, so we cannot account for any minor or moderate traumas that may have occurred. We may have underestimated the true rate of major trauma recidivism, as some patients grouped as non-recidivists could have incurred another major trauma outside the province or beyond the study period. While alcohol status was categorized as positive or negative, our 2 mmol/L cut point is considerably lower than the forensic/legal limit of 17 mmol/L. Further research is warranted to examine whether the degree of intoxication has an impact on major trauma recidivism.
CONCLUSION
In this provincial study of major trauma recidivism, 2% of patients were recidivists, but their likelihood of mortality was over three times greater, as compared with non-recidivists after adjusting for potential confounders. Screening and brief intervention may help to reduce recidivism rates.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/cem.2018.496.
Acknowledgements
The data used in this research were made available by Trauma Nova Scotia at the Nova Scotia Department of Health and Wellness. Any opinions expressed by the authors do not necessarily reflect the opinion of the Nova Scotia Department of Health and Wellness or Trauma Nova Scotia. This work was supported by a Clinician Scientist Award from the Faculty of Medicine, Dalhousie University, Halifax, NS, Canada.
Competing interests
None declared.
