Abstract
Objectives:
We sought to evaluate the impact of race on treatment approaches and mortality following arterial trauma.
Methods:
The National Trauma Data Bank (version 7.2, American College of Surgeons) was queried from 2002 to 2012 to identify patients aged 18 to 65 years with arterial trauma. The association between race (white, black, and Hispanic) and mortality following arterial injury was assessed, stratified by penetrating or blunt injury. Temporal trends in the use of open and endovascular procedures were evaluated across the racial groups. Multivariable regression models adjusting for patient demographics, injury severity, hospital characteristics, insurance status, and type of intervention performed were used to evaluate potential contributors to the association of race with mortality.
Results:
The study cohort consisted of 58 626 patients (52% white, 31% black, and 17% Hispanic). A majority (57%) of patients had penetrating injuries, with black and Hispanic patients being more likely to sustain penetrating injuries (80% and 65%, respectively) compared to white patients (41%, P < .001). Overall, black patients had higher mortality for penetrating injuries (16.8% vs 13.0% vs 7.8%, P < .001) when compared to Hispanic and white patients, correspondingly. Over the study period, there was increasing use of endovascular and decreasing open surgical procedures for treatment of arterial trauma. This finding was similar across all groups studied. In multivariable analysis, black race was found to be associated with higher mortality compared to white for both penetrating (odds ratio [OR] 1.52, 95% confidence interval [CI] 1.33-1.75, P < .001) and blunt (OR 1.27 95%CI 1.09-1.47, P = .002) arterial trauma.
Conclusion:
Even after adjusting for potential confounders, minority patients had increased odds of mortality following arterial trauma compared to their white counterparts. Further studies are needed to understand and to eliminate these observed disparities in outcome.
Introduction
Racial and ethnic differences in health care outcomes have been widely studied and reported. 1 -9 Trauma, which disproportionally affects minorities, has often been cited as one of the major contributors to this observed phenomenon. 10 -16 Black and Hispanic patients have been reported to have higher injury mortality rates than whites even when patient and societal factors such as injury severity, insurance status, and hospital characteristics are controlled for. 13,17,18 Some investigators have alluded to differences in baseline health and genetic variations in disease processes without fully explaining the underlying causes of these differences. 6,8 In as much as trauma continues to be a leading cause of morbidity and mortality for adults in the United States, continued efforts should be put forth to identify and reduce unwanted disparities in observed outcomes.
In the vascular literature, studies examining race-specific difference in outcomes have produced mixed results. Black patients have been noted to have worse outcomes following carotid endarterectomy, peripheral artery bypass grafting, and abdominal aortic aneurysm repair. 18 -23 In a study examining lower extremity vascular injuries, mortality disparities based on race were only observed for penetrating trauma. 24 In a large Veterans Administration study of vascular procedures, however, race was not found to be a significant predictor of outcome. 25
The role of race in arterial trauma has not been well described. In the present study, we evaluated the impact of race on in-hospital mortality and treatment approaches for arterial trauma in the National Trauma Data Bank (NTDB), the largest national database of patients with trauma. The approach to care in vascular trauma continues to undergo dramatic changes with increasing use of endovascular therapies. We speculate that differences in the application of these newer techniques across racial groups may help to partially explain some of the differences in outcomes. Therefore, we also examined whether open and endovascular procedure use differed among the groups studied in order to determine the trends in mechanism of care and utilization of endovascular therapy for the treatment of arterial trauma.
Methods
This study is a retrospective review of prospectively collected data from 2002 to 2012 in the NTDB, which is managed by the American College of Surgeons (ACS) Committee on Trauma (COT). The ACS established the NTDB as a public service to be a repository of trauma-related data voluntarily reported by participating trauma centers. Currently, the NTDB is the largest repository of data on patients with trauma in the United States, containing detailed information on over 4 million cases from over 900 trauma centers. 26 The data set collected by NTDB is considered a limited data set under Health Insurance Portability and Accountability Act since the research data set that ACS releases is deidentified. Informed consent was waved for this study, and final exemption was provided by the Institutional Review Board at Boston University School of Medicine.
Adult patients aged 18 to 65 years with traumatic arterial injuries to the neck, thorax, abdomen and pelvis, upper and lower extremities, and no head trauma based on International Classification of Diseases, Ninth Revision (ICD-9 codes; Appendix A) were included. Figure 1 shows the flowchart for the patient selection process for our study. Patients with combined penetrating and blunt injuries, those dead on arrival to the emergency department (ED), and lacking discharge status were excluded. We also excluded cases without gender and race designation in the data fields and those with race designations other than white, black, or Hispanic. This resulted in approximately 28% of eligible patients being excluded because of missing elements of data. Comparison of racial groups was the main aim of the study.
Flowchart of the patient selection process for the study.
We first compared patient demographic characteristics including age, gender, the presence of comorbidities, and insurance type. Then, we evaluated injury characteristics including mechanism of injury (penetrating or blunt), injury severity score (ISS), systolic blood pressures (SBP) in the ED, and anatomic location of injury. Additionally, we compared hospital characteristics such as type of hospital (university vs nonuniversity), hospital COT designation (Level 1 vs other), and geographic region. A bivariate analysis using a chi-square test or t test was implemented to compare these characteristics.
To study linear time trends in utilization of open and endovascular procedures by injury type across 3 racial groups, we used logistic regression with race, year, and Race × Year interaction terms. End points analyzed included interventions performed, hospital lengths of stay (LOS), and mortality. Analysis of categorical variables was performed using the chi-square test, and an unadjusted γ regression was used to compare the overall LOS in the groups. For the multivariable analyses, we stratified the cohort by mechanism of injury (penetrating vs blunt). For the main end point of mortality, sequential multivariable logistic regression models were constructed to adjust for age, gender, ISS, insurance type, geographic region, complications, and procedures. For all tests, the type I error level was set at 0.05. All analyses were performed using SAS 9.2 (SAS Institute Inc, Cary, North Carolina).
Results
As shown in Table 1, a total of 58 626 patients met the study inclusion criteria and were included in the analysis. There was broad racial representation with 31% black patients and 17% Hispanic patients. Several characteristics varied across the racial groups including demographics (age and sex), injury characteristics (mechanism, location, ISS, and ED SBP), insurance status, and hospital characteristics (Level 1 Trauma Center, hospital affiliation).
Demographic Characteristics of Patients With Arterial Trauma by Race/Ethnic Group.
Abbreviations: ISS, injury severity score; SBP, systolic blood pressures; ED, emergency department; SD, standard deviation; COT, Committee on Trauma.
Overall, the study cohort consisted of moderately injured patients with a median ISS of 11 to 13. A majority of the patients were stable on arrival to the ED with mean SBP of 118 ± 38.1 mm Hg. Of those patients who did present with hemodynamic instability, defined as SBP <90 mm Hg, a higher proportion were black (21%) or Hispanic (17%) compared to white (14%, P < .001).
Table 2 provides overall mechanism of injury, interventions, and crude outcomes stratified by racial group. Black patients with arterial injury had a higher prevalence of penetrating injury (80%, P < .001) compared to their white and Hispanic counterparts (41% and 65%, respectively). Overall patterns of the use of open and endovascular procedures for arterial trauma differed across the racial groups. Black and Hispanic patients had higher rates of open surgical procedures and lower rates of endovascular interventions. In analysis stratified by mechanism of injury (penetrating vs blunt), the racial differences in treatment approaches persisted (Figure 2). For penetrating trauma, black and Hispanic patients had higher rates of open procedures and lower rates of endovascular procedures compared to white patients. In the setting of blunt arterial trauma, Hispanic patients had higher rates of open procedures and lower rates of endovascular interventions.
Injury Mechanism, Interventions, and Unadjusted Outcome of Patients With Arterial Trauma by Race/Ethnic Group.
Abbreviations: LOS, length of stay; SD, standard deviation.
Intervention type for penetrating and blunt arterial trauma across race/ethnic group.
Endovascular procedures are increasingly used to treat arterial diseases. Temporal trends of endovascular and open procedure use for arterial trauma across the racial groups are shown in Figure 3 separately for penetrating and blunt trauma. For penetrating arterial trauma, there was an increasing use of both endovascular and open procedures over the time period. For blunt arterial trauma, there was an increasing use of endovascular procedures and decreasing use of open procedures. The temporal trends were similar across the racial groups.
Temporal trends in utilization of open and endovascular intervention for blunt and penetrating injury across race/ethnic groups.
Despite the differences in mechanism of injuries and interventions provided, the overall hospital LOS was similar across the racial groups (Table 2). When we stratified LOS by injury mechanism (penetrating vs blunt), we found that patients who had penetrating injuries had shorter LOS than those with blunt injuries. White patients had shorter LOS for penetrating injuries (7.1 ± 11.5, P < .001) but not for blunt injuries (12.8 ± 15.8, P = .224) compared to blacks (penetrating 9.7 ± 16.2; blunt 13.1 ± 18) and Hispanics (penetrating 8.9 ± 14.9; blunt 12.3 ± 16.3). However, whites also have significantly higher proportion of blunt injuries compared to blacks and Hispanics, thereby increasing the mean of the overall LOS leading to the nonsignificant racial difference in combined sample.
Overall, black patients had higher mortality rates (16%, P < .001) compared to white and Hispanic (10% and 12.7%) patients (Table 2). As shown in Figure 4, when evaluated by mechanism of injury, black patients had significantly higher mortality for penetrating trauma when compared to Hispanic and white patients. Black race was associated with higher mortality rates for blunt arterial injuries as well, although this difference did not reach statistical significance with a P value of .077.
Racial disparities in mortality following arterial trauma by mechanism of injury.
As there were multiple injury and patient characteristics that differed by race, we used multivariable models to evaluate potential contributors/confounders of the association between race and mortality for penetrating or blunt arterial injury. Shown in Figure 5 are the results from sequential logistic regression models, adjusting for demographics (age and sex), injury severity (ISS and ED SBP), hospital characteristics (region, University affiliation, and Level 1 trauma designation), insurance status, injury characteristics (anatomic location), and treatment approach (endovascular or open).
Adjusted odds of mortality by race/ethnic group for penetrating and blunt arterial trauma*.
For penetrating arterial trauma, the association of Hispanic race with mortality was rendered not significant after adjusting for hospital characteristics. In contrast, the association of black race with higher mortality was only minimally attenuated in the fully adjusted model suggesting that the potential confounders/mediators only partially accounted for the association of black race and higher mortality for penetrating trauma.
For blunt arterial injury, multivariable adjustment attenuated the association of Hispanic race with mortality. However, the association of black race with higher mortality persisted with a similar odds ratio. In both penetrating and blunt vascular trauma, adjusting for treatment approach did not alter the association with mortality suggesting that racial differences in treatment utilization do not account for differences in mortality.
Discussion
Our study shows that there are racial disparities in mortality outcomes following arterial trauma. We delineated key temporal trends demonstrating the adoption of endovascular therapies for both penetrating and blunt arterial injury. Importantly, although there were differences in treatment approaches by race, temporal trends were similar across racial groups. Hispanic patients had higher mortality for penetrating and blunt trauma that was accounted for by differences in demographics, injury severity, and hospital characteristics. In models accounting for multiple factors, black patients had 27% higher mortality rates for blunt and 52% higher mortality for penetrating arterial trauma when compared to white patients. Procedure utilization did not appear to account for the higher mortality risk in black patients for either penetrating or blunt trauma indicating that access to new treatments does not appear to underlie racial disparities. Taken together, our findings suggest a persistently higher risk of death in black patients with arterial trauma not fully accounted for by differences in injury severity or location, insurance or hospital status, and treatment approach.
The presence of racial disparities in our study is in keeping with a larger body of literature emerging in the past several years noting that outcomes after trauma are not “color blind.” 13 -15 We hope to highlight the need for continued research to identify, measure, and improve the quality of trauma care received by all patients regardless of race. Recent estimates suggest that 5% of deaths from traumatic injury are associated with racial disparities in care. 27 Data from the trauma literature and other areas of health care that have examined racial disparities often cite poor access to care, lower socioeconomic status, preexisting medical conditions, and potential provider biases as some of the reasons for these findings. Prior studies have suggested that black patients may receive their care at hospitals with few resources to provide high-quality care, and hospitals with poorer outcomes receive care from lower volume, lower quality surgeons compared to white patients with trauma. 14-15,19 Haider et al reviewed data from the NTDB and demonstrated that moderate to severely injured patients who were treated at predominately “minority hospitals” where >50% of patients were minorities had a 37% higher risk of mortality than those treated at predominately “majority hospitals” even after adjusting for the aforementioned confounders. 15
Our study extends the prior work by focusing specifically on arterial trauma. Similar to prior studies, we observed racial differences in injury mechanism and severity as well as insurance status and hospital type. Black patients were more likely to be taken to a University hospital and Hispanic patients more likely to be treated at COT Level 1 trauma centers. However, multiple indicators related to injury status, access to care, and hospital characteristics did not fully attenuate the association of black race with higher mortality following arterial injury. Interestingly, Frankema et al found that the use of ISS and other injury scores have not been shown to adequately describe severity of penetrating injuries. 28 Thus, it remains possible that the degree of difference in injury severity was not fully described in the available metrics in the NTDB collection. Further, socioeconomic status is not well captured in NTBD and has been shown to relate to racial disparities in trauma outcomes. 29 Alternatively, additional factors reflecting access to care or provider behavior may be responsible for the markedly higher rate of dying from arterial injury in black patients. 30
Given the dramatic changes in the treatment approaches to arterial diseases, we focused on procedure utilization for arterial trauma as a potential factor contributing to racial disparities. Approach to care in vascular trauma has undergone dramatic changes over the last decade, with increasing use of minimally invasive endovascular therapies. In particular, endovascular therapy is increasingly used to treat blunt aortic trauma. Additional studies have expanded the application of endovascular repair beyond thoracic injuries in the adult population. 31 -35 In our study, we sought to better quantify the expanding role of endovascular technology in arterial trauma. Overall, there were differences in treatment utilization for both blunt and penetrating arterial trauma across the racial groups. There were important ongoing temporal trends with increasing endovascular therapy utilization for both blunt and penetrating arterial trauma. 36 Interestingly, when evaluated across the racial groups, the rate of adoption of endovascular therapy appeared similar for black and Hispanic patients when compared to white patients for both blunt and penetrating injuries. Thus, disparities in access to new technologies do not appear to explain racial differences in outcomes in arterial injury.
This study had several important limitations. Our analysis relied on data from the NTDB, a prospectively collected database that is subjected to both selection and information biases due to potential differences in reporting between different institutions. Because the NTDB is not a population-based sample of hospitalized patients and includes a disproportionate number of larger hospitals, with younger and more severely injured patients, the findings may not be fully representative of the national experience. Furthermore, the use of ICD-9 codes to select patients has limitations due to reliance on physician-documented diagnoses and may not fully capture injury complexity. There is likely residual confounding due to unmeasured covariates including the time elapsed between injury and hospital presentation, and pre- and posthospitalization management or outcome information. We did not evaluate patients older than 65 years, and the racial associations may differ in this age-group. 37 The shortcomings are counterbalanced by the large sample size, racially representative population, and long time period of data collection.
Conclusion
Black and Hispanic patients have worse outcomes with higher mortality rates after arterial trauma compared to their white counterparts. For black patients, the racial disparities in mortality are not fully explained by demographics, insurance, or injury severity. Although there were differences in procedure utilization by race, procedure use did not account for racial disparities. Further, important temporal trends in endovascular and open surgical treatment utilization for arterial trauma were similar across racial groups. Further studies are needed to better understand and eliminate racial disparities in arterial trauma outcomes.
Footnotes
Appendix A
ICD-9 Codes for Arterial Injury by Anatomic Distribution Used in the Study.
| Anatomic region | Codes |
|---|---|
| Neck | 900.0, 900.00, 900.01, 900.02, 900.03, 900.1, 900.8, 900.81, 900.82, 900.89, 900.9 |
| Thorax | 901.0, 901.1, 901.2, 901.3, 901.40, 901.41, 901.42, 901.81, 901.82, 901.83, 901.89, 901.9 |
| Abdomen and pelvis | 902.0, 902.10, 902.11, 902.19, 902.2, 902.20, 902.21, 902.22, 902.23, 902.24, 902.25, 902.26, 902.27, 902.29, 902.31, 902.32, 902.33, 902.34, 902.39, 902.4, 902.40, 902.41, 902.42, 902.49, 902.50, 902.51, 902.52, 902.53, 902.54, 902.55, 902.56, 902.59, 902.8, 902.81, 902.82, 902.87, 902.89, 902.9 |
| Upper extremity | 903.01, 903.02, 903.1, 903.2, 903.3, 903.8, 903.9 |
| Lower extremities and unspecified | 904.0, 904.1, 904.2, 904.3, 904.4, 904.40, 904.41, 904.42, 904.5, 904.50, 904.51, 904.52, 904.53, 904.54, 904.7, 904.8, 904.9 |
Abbreviations: HIPAA, Health Insurance Portability and Accountability Act; ACS, American College of Surgeons; SD, Standard deviation; COT, Committee of Trauma; ISS, Injury severity score; ED, Emergency department; SBP, Systolic blood pressure; LOS, Length of stay; ICD-9, International Classification of Diseases, Ninth Revision.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.