Introduction

Racial disparities in the quality of and access to health care have been revealed in almost all areas of medicine. There is conflicting evidence on whether fewer racial disparities exist in emergency care as compared to elective and preventive care [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17]. The management of injured patients, which accounts for a substantial component of emergency care and affects patients from all racial/ethnic and socioeconomic backgrounds, has become increasingly protocolized over the last three decades. While mortality rates for severely injured patients have decreased in recent years [18], evidence of racial disparities in trauma care continues to emerge. It was recently estimated that, over a 4-year period, approximately 5% of trauma deaths were attributable to racial disparities in trauma care [19••].

Previous reviews of the literature have shown that race/ethnicity and insurance are associated with disparate outcomes following trauma [20,21,22]. Several studies have illustrated that Black patients have higher mortality rates than comparably injured White patients, partially due to the fact that Black patients are more likely to have penetrating mechanisms of injury [23,24,25,26,27,28,29]. Mortality rates differed when grouping all non-White patients together; some studies found that non-White race was a predictor of mortality in trauma [23, 30], while others showed no difference. With regard to long-term functional outcomes after traumatic injury, however, it is apparent that all racial minority groups fare worse than White patients [5, 31,32,33,34,35].

Studies have also shown that lack of insurance is a predictor for adverse outcomes in trauma surgery [26, 36,37,38,39,40,41,42]. Haider et al. demonstrated that insured patients had lower crude mortality rates than similarly injured uninsured patients, but Black and Hispanic patients had worse outcomes than White patients, regardless of insurance status [9]. When comparing blunt and penetrating trauma, lack of insurance is a predictor of mortality among similarly injured patients [28]. The deleterious effect of being uninsured on outcomes in trauma have been emphasized in the pediatric trauma literature, with numerous studies demonstrating that uninsured and publicly insured children have higher mortality after sustaining trauma [27, 43]. As the Centers for Disease Control (CDC) has identified expansion of access to trauma care as a Healthy People priority [44], a thorough examination of these disparate outcomes is necessary to improve short-term and long-term outcomes for patients of color/racial minority patients following traumatic injury.

The literature on quality measurement and performance improvement in trauma care has also expanded in parallel with the literature on trauma disparities. In 2008, the American College of Surgeons (ACS) established the Trauma Quality Improvement Program (TQIP) to benchmark trauma centers. Several studies have since demonstrated a large degree of variability in hospital-level performance on both process and outcome measures [45,46,47]. There is evidence that patient selection and case mix, including racial and socioeconomic characteristics, contribute in part to this variability [20, 48, 49]. This evidence, which is summarized in detail in this review, is of critical importance to the development of an understanding of the current state of healthcare disparities and quality measurement in trauma surgery.

Objective

The objective of this review is to critically assess and summarize racial disparities in trauma outcomes, both short term and long term, described in the current body of literature (encompassing the last 5 years), in the context of recent developments in trauma system organization, health policy, economic infrastructure, and surgical education.

Methods

We searched the electronic database PubMed and the Cochrane Library. The search criteria, which were restricted to primary research articles published from 2012 to 2017, included the keywords “health disparities,” “emergency,” “surgery,” “quality improvement,” and “ethnic groups,” with “trauma” included as a “title/abstract” search filter term. A total of 47 articles met the search criteria for this review, as shown in Table 1. Studies in which surgical patients of all ages were included, including pediatric and geriatric populations, and the outcomes in the studies varied from short-term outcomes such as in-hospital mortality to longer term outcomes, such as discharge to rehabilitation facilities. Studies examining process measures were sought by adding “process” to the search as a title/abstract filter term. In the absence of randomized controlled trials, our review focused primarily on retrospective and prospective cohort studies. The review was performed in a systematic manner, beginning with the broadest search and progressively adding search and filter terms to allow the development of a comprehensive analysis of the most current literature in trauma disparities.

Table 1 Annotated table of references
Full size table

Summary of Recent Findings in the Trauma Disparities Literature

As process measures are increasingly used to measure clinical performance in trauma, disparities in the adherence to evidence-based protocols represent an important variable in the study of trauma disparities. Efforts to improve quality of care for geriatric trauma patients, for instance, have focused on several process-oriented quality indicators [96,97,98]. At the moment, no such metrics exist for benchmarking trauma centers according to racially disparate outcomes and processes. Synthesizing the data from the trauma quality improvement literature with the disparities literature to establish methodologies for measuring disparities as a quality indicator is an essential future direction to pursue if the discipline of trauma care is to continue to move forward.

Recent investigations into disparities in trauma have expanded upon the existing body of knowledge by not only revealing disparities that had not been previously characterized but also reflecting the shift in focus from immediate outcomes to longer term functional outcomes and systems interventions. Post-injury patient-reported quality of life has emerged in parallel to this trend as an outcome measure of care quality. Additionally, a number of studies described age-specific trends in racial disparities among injured patients, particularly in older age groups. As process measures evolve to track quality of care for geriatric trauma patients, racial disparities among this vulnerable population take on a new dimension of significance. Finally, a recurring theme in the updated body of trauma disparities literature is the effort to understand the complex mechanisms underlying racial and socioeconomic disparities in quality of trauma care. These findings are described in detail below.

New Disparities in Trauma Surgery

The early 2000s saw a wave of studies that defined disparities in various aspects of trauma care. This trend continued over the past 5 years, during which numerous studies have emerged detailing disparities in the management of injured patients that had not been previously described. Black patients, especially those without insurance, are more likely to die following an intentional injury, with uninsured Blacks comprising 76% of all excess trauma center deaths due to assault [55]. Traumatic brain injury (TBI) among veterans, both combat and non-combat, was found to differentially affect patients based on their race; one study found that Hispanic ethnicity was a predictor of mortality following TBI [51] while another found that both Black race and Hispanic ethnicity were associated with more severe TBI [78]. The authors postulated the differences in adherence to numerous processes of care that inform outcomes in TBI as potential contributors to these disparities.

Poor outcomes among injured patients are associated with non-White race, according to several recent studies. Adult patients with traumatic spinal cord injuries experience different outcomes depending on their race, with Black and Hispanic patients demonstrating longer risk-adjusted length of stay than White and Asian patients. Interestingly, the same study also found that Black and Native American patients have higher odds of complications than White, Hispanic, or Asian patients [70]. Similarly, Schoenfeld et al. found that non-White patients have an increased mortality risk following spine trauma, a relationship that is amplified by lack of insurance [62•, 91]. These disparities persist among children as well; in a nationwide sample of pediatric patients with spinal injuries, Black children were more severely injured and were less likely to undergo definitive operative intervention [81]. Black patients who suffer traumatic amputation of the digits, an injury associated with devastating consequences to quality of life and vocational performance, are less likely than White patients to undergo replantation, despite advancements in microsurgical techniques [85]. They are also less likely to receive surgical intervention for facial fractures [95]. Among patients requiring emergency laparotomy and fecal diversion due to hollow viscus injury, Black and Hispanic patients were less likely to ultimately undergo stoma reversal procedures than their White counterparts [99]. The authors speculated that this was due to the increased likelihood of Black and Hispanic patients lacking insurance, rather than any hospital-level quality indicators, but acknowledged that this conclusion is difficult to defend in the absence of validated process measures for this condition. Black race is also a strong predictor of developing sepsis [68] and acute kidney injury [69] among severely injured patients, which others have attributed to the fact that Black patients tend to cluster at poorer-quality hospitals [59••].

Insurance disparities, which are distinct from but closely related to racial disparities, have also been evaluated and defined in the context of trauma in recent years. Uninsured patients who sustained penetrating trauma are more likely to have shorter ICU lengths of stay and longer overall hospital lengths of stay [52] and less likely to receive inpatient procedures while hospitalized for TBI [90]. Those with pelvic fractures are less likely to receive clinically indicated diagnostic tests and therapeutic interventions, such as embolization, and even have higher mortality rates [65]. Serious complications, such as anastomotic leaks and fascial dehiscence, are more common in uninsured patients who sustain traumatic hollow viscus injuries [71]. Another nationwide study demonstrated that inpatient mortality following traumatic injury is higher among lower income patients, even after risk adjustment, and uninsured trauma patients who present without vital signs and undergo urgent thoracotomy/laparotomy are less likely to survive than insured patients [66, 77]. Insurance disparities in a relatively new quality metric known as failure to rescue have also been identified, with uninsured patients being more likely to experience failure to rescue than publicly or privately insured patients [64•]. Many of the authors of these studies have sought to explain these associations by theorizing that uninsured patients tend to cluster at trauma centers that perform below average on both process and outcome measures.

One study argued that Black race and/or Hispanic ethnicity were not associated with worse outcomes in trauma patients. The authors extrapolated from a nationally representative sample that there was no difference in risk-adjusted survival among injured patients based on race or insurance status [80]. However, other investigators disputed these conclusions on the grounds that the data set used to perform this analysis lacked the clinical granularity needed to meaningfully interpret data on trauma outcomes. Therefore, adjusting for known confounders such as injury severity and trauma center level designation was rendered impossible by the inherent limitations of the selected data set. Furthermore, all the aforementioned studies that did find racial disparities restricted the sample to those with moderate to severe injuries, whereas this study alone included patients with mild injuries who are at low risk for mortality, resulting in attenuation of the observed effect [100] (Fig. 1).

Fig. 1
figure 1

The interaction among underrepresented minority race/ethnicity, young age, and lack of insurance coverage can potentiate disparities in trauma

Full size image

Age-Related Disparities in Trauma Surgery

In the management of trauma, patient age is associated with differences in injury mechanism, injury severity, clinical outcomes, and functional outcomes; these differences are further influenced by patient race and insurance status. Hicks et al. described a paradoxical association between race and mortality among elderly trauma patients; in a large nationwide sample, older Black patients actually had reduced mortality compared to similarly injured older White patients, in contrast to the higher risk-adjusted mortality risk among young Black trauma patients [73•]. Further study by this group revealed that young Black trauma patients tend to receive care at lower performing trauma centers, whereas older Black patients were less likely to be treated at these poorer-quality facilities, suggesting an important age-related disparity in the management of injured patients [75]. Another study asserted that among elderly trauma patients, only Hispanic ethnicity is a predictor of increased mortality. Older patients in this cohort are also less likely to be uninsured than their younger counterparts [61]. Among patients with isolated vascular trauma, however, race is the single greatest predictor of adverse outcomes in the elderly, with older Black patients being five times more likely to experience death or amputation [82].

On the other end of the age spectrum, injured children also experience age-dependent disparities, although insurance appears to be an effect modifier for racial disparities in the pediatric population. Having insurance is associated with a reduction in mortality in a nationwide sample of injured children, as adjusting for insurance mitigates the racial disparities seen in this population [63•]. A similar phenomenon was described in a state database, which showed that outcomes differ according to race on unadjusted analysis, but adjusting for variables that determine access to care explains and eliminates these disparities [67]. Multiple studies using data from institutional trauma registries found that Black race was associated with lack of insurance, higher mortality rates, fewer diagnostic tests, and higher rates of penetrating injury in a cohort of critically ill pediatric trauma patients, which suggests that the disparities seen in this population are multifactorial in etiology [53, 54, 81, 83, 92, 101, 102]. As there are currently no validated quality metrics for access to trauma care, these new data indicate that there may be a fundamental difference in baseline health status for uninsured children, as well as children belonging to underrepresented minority groups, that puts them at increased risk for unfavorable outcomes following traumatic injury, but is not captured by current quality measurement techniques.

Pre-Hospital and Post-Hospitalization Disparities

Fortunately, despite the limitations of measuring quality as a function of access to care, our understanding of disparities in trauma care now extends beyond in-hospital outcome measures to those that occur before and after hospitalization. For instance, one study described a pre-hospital disparity in trauma that was previously unexplored: the use of restraints in motor vehicle collisions and its relationship to outcomes. The authors of this study found that Black, Hispanic, and Native American children are the least likely to use restraints and have more severe injuries [56••]. Another study found that among children who sustained injuries as a result of bicycle accidents, Black children were less likely than White children to be wearing a helmet at the time of the accident [103] White patients are more likely than Black or Hispanic patients to receive pre-hospital analgesia for blunt trauma injuries [57]. Research in pre-hospital disparities has also uncovered geographical disparities in trauma care; patients in rural areas, for example, are more likely than patients in non-rural areas to die after traumatic injury, thought to be the result of increased travel time for rural patients to receive treatment [86, 104]. While not a racial disparity per se, geographic disparities represent an important public health concern that is likely to disproportionately affect the most socioeconomically disadvantaged patients.

Assessment of functional outcomes after trauma has also come to the forefront in recent years, and racial disparities exist in the post-hospitalization phase of care as well. For patients who sustain falls, the risk of post-discharge mortality appears to be higher among Blacks [84]. Black and Hispanic patients are less likely to be discharged to post-acute inpatient rehab after traumatic brain injury, even when insured [60, 74]. Racial minorities and patients with Medicaid as their primary insurance payer have poorer long-term functional outcomes following lower extremity fractures that are treated operatively [93]. Clinical decision-making surrounding end-of-life care also varies across trauma patients of different races, with Black patients receiving more escalation of care and Hispanic patients being less likely to have withdrawal-of-care orders, indicating the presence of heretofore unexplored cultural preferences and interactions with the healthcare system among different racial groups [105]. Furthermore, disparities in post-discharge rehabilitation referrals following traumatic brain injury have been observed for racial/ethnic minorities, as well as those without insurance [76, 87, 88, 94].

Finally, unplanned readmissions after receiving treatment for traumatic injuries represent another quality measure for which minority patients have inferior outcomes. Black race was found to be a predictor of unplanned 30-day readmission following trauma [89, 106].

Causal Pathways of Disparities in Trauma

Researchers have focused on elucidating the mechanisms underlying trauma disparities in recent years. Fundamental differences in the characteristics of the hospitals at which racial minority patients tend to receive their trauma care have been identified as an important contributor to disparate outcomes in this population. Patients treated at hospitals at which racial minority patients account for greater than 25% of the population have increased odds of death and major complications [50•, 58], and Black patients are more likely than Hispanic or White patients to be treated at these high-mortality centers [59••].

Other theories to explain disparities in trauma include the relationship between minority race and lack of insurance or inadequate insurance. Recent studies have shown that patients without insurance or with Medicare have higher odds of death following blunt injury than patients with Medicaid [107]. Uninsured patients are also less likely to have diagnosed comorbid conditions at the time of injury, yet still have higher odds of mortality than their insured counterparts, suggesting that the burden of undiagnosed comorbidities in this population is an underlying factor in the observed mortality difference [108]. Policy initiatives aimed at expanding insurance coverage have disproportionately benefited White patients in South and West census regions, with young Black and Hispanic trauma patients experiencing similar uninsured rates before and after the introduction of legislation [79••].

One study sought to examine the provider-level behaviors that could influence disparities in trauma, namely, unconscious race and class bias. In the study sample, a large majority were found to have implicit biases towards upper socioeconomic classes and in favor of White patients, but these biases were not associated with vignette-based clinical decision-making. Whether unconscious bias affects clinical decision-making outside of simulated scenario has not yet been established, this area represents a new frontier of research in trauma disparities [72].

Study Limitations

This evidence-based review of the current literature on racial disparities in trauma surgery, while illustrative of the extant knowledge in the field, is limited by the absence of experimental studies. The preponderance of observational studies in the disparities literature is not a recent phenomenon; previous reviews on this topic have also yielded predominantly retrospective cohort studies that used administrative data sets. Research in healthcare disparities is challenging to accomplish within the constraints of a randomized controlled trial, for practical and ethical reasons, and there are few proposed interventions for reducing disparities that lend themselves to investigation in this format.

Conclusions

Understanding the scope of the problem of racial disparities in the management of injured patients is an imperative research objective in the overall effort to eliminate healthcare inequities. Investigations over the last 5 years have expanded upon this objective by defining disparities in populations at the extremes of the age spectrum and in the pre- and post-hospital phases of care and by exploring the underlying causes of previously defined disparities in trauma. Furthermore, there is a deficit in the ability of existing quality metrics to capture disparities in trauma processes and outcomes. Although the literature has yet to describe and test definitive interventions for reducing disparities in access to and quality of trauma care, this review indicates that research in racial disparities in trauma is evolving towards a more comprehensive understanding of these inequities and the approach that future investigators may ultimately take to mitigate their effect on vulnerable patients.