Factors associated with hospital mortality in traumatic injuries: Incentive for trauma care integration

Summary

Objectives

The main aim of this study was to contrast the variation in mortality between trauma centres (TCs) and non-trauma hospitals (NTHs) in Texas, and among TCs by sociodemographic and economic factors of trauma cases.

Study design

Difference in fatality due to trauma by hospital type was studied for all injured cases hospitalized over a 2-year period.

Methods

The outcome measure was mortality following an injury for cases that survived the impact and were treated in any hospital. Logistic regressions were employed to compare the risk factors associated with trauma fatalities between TCs and NTHs, and among TCs.

Results

The risk of dying at a TC in contrast to an NTH was higher among young adult males and cases admitted through the emergency department/room. In rural areas, fatality was higher among 25–44 year olds, Hispanics, uninsured patients, and cases admitted through transfer. In urban settings, fatality was higher among 18–24 year olds, patients covered by ‘other’ insurance, and cases admitted as severe emergencies. Increased mortality at Level I TCs occurred due to the transfer of patients from rural areas. Blacks and Hispanics in rural areas were more likely to die, while Hispanics had lower fatality in Level I TCs in urban areas. Survival time was longer for patients treated in urban TCs compared with rural TCs.

Conclusion

In the absence of validated data about severity of cases and type of injury, and details about the treatment provided to trauma cases in this study, more investigation is needed into the case-mix of trauma patients admitted to TCs and NTHs. Further exploration is necessary for better co-ordination of the emergency care response to integrate NTHs within the trauma system and alleviate the stress placed on Level I TCs. Revisiting the transfer algorithms could improve clinical outcomes, particularly when TCs are closed due to diversion protocols.

Introduction

A nationwide inventory of trauma centres (TCs) showed the distribution and coverage of trauma care as part of an evaluation to design an appropriate response to possible mass casualties, and compared their availability with that of acute hospitals in general.¹

With significant variations found between states in capacity to respond to major crises, emphasis was placed on the implications for equity, effectiveness and efficiency of trauma services, and the need for additional information at state level to influence health policy and address geopolitical disparities in the distribution and coverage of TCs.²

The inventory identified Texas among the states with well-developed trauma systems and as the state with the largest number of facilities.¹ Following publication of the inventory, the authors undertook an investigation to compare the characteristics of patients with fatal traumatic injuries treated at TCs in contrast with those treated in non-trauma hospitals (NTHs) in Texas.

This investigation was concerned with trauma fatalities at all hospitals that treated injured patients in Texas; one of the largest and most rural states in the USA. Two specific aims were pursued in this study. Firstly, to establish the profile of risk factors associated with trauma mortality at TCs compared with NTHs. Specifically, injury mortality was compared for TCs and NTHs statewide and between rural and urban areas to document variation in the characteristics of the population of patients who died in both types of hospital following a traumatic injury. Secondly, differences in fatalities among TCs were analysed by their designated capacities (Levels I–IV) to profile the risk of dying at each type of facility.

Designation of TCs is based on the extent to which each centre meets essential or desired criteria regarding organization, clinical qualifications, available resources, rehabilitation services, performance improvement, continuing education, prevention and research, with Level I being the most comprehensive.³ Examining the case-mix of trauma fatalities and the related risk factors of differential mortality at TCs compared with NTHs and among different levels of TCs can potentially provide an estimate of the burden placed on these hospitals by trauma cases. Specifically, potential variation in outcomes at the two types of hospital in relation to sociodemographic and economic factors represented the focus of this investigation.

With very few exceptions, most previous studies concerned with trauma care have used mortality as an outcome measure. When reviewing the studies of differential mortality due to trauma at TCs and NTHs, the present authors’ assumption was that all states applied similar methods to verify and classify a hospital as a TC. Further, when examining the literature, two groups of studies were distinguished. One group was primarily concerned with presenting mortality and the associated risk factors at TCs exclusively, and the other group compared the variation in outcomes between TCs and NTHs.

Several investigations found that increased age was a risk factor for higher mortality.4, 5, 6, 7, 8 Vulnerability of older patients to trauma led some researchers to focus specifically on patients over 65 years of age.7, 9, 10 Other demographic factors associated with higher mortality were ethnicity9, 11 and gender.5, 9, 12 Specifically, Blacks were found to be more likely to die at Level I TCs,⁹ while Whites had an increased case mortality rate at Level II centres.¹¹ Overall, males had a higher risk of death following traumatic injuries and spent more days in hospital than females in all age groups.¹² Lack of insurance was also found to be associated with higher mortality among hospitalized cases after an injury.¹³

Previous studies found that increased mortality was associated with higher severity of injury,5, 7, 8, 9, 10 mainly related to multiple organ failure, a longer stay in intensive care units,¹² and the number of injuries and complications due to comorbidities.⁷ Mortality was also higher for patients who needed transfer within the trauma system.6, 14 The average length of hospital stay was correlated with lower mortality, especially for the most severe patients.5, 15, 16

Similar studies to this have compared the outcomes between TCs and other acute hospitals at local and national levels.1, 4, 6, 8, 10, 13, 15, 17, 18, 19 The findings of these investigations present a rather unclear picture with inconsistent conclusions. An early investigation found that most patients who died at TCs were aged 45–65 years old, male and had shorter hospital stays, while patients who died in NTHs were mainly over 65 years of age, female, and had longer hospital stays.⁴ Mortality rates were higher among elderly patients at NTHs compared with TCs, and were associated with injury severity in one county.¹⁰ Although uninsured injured patients were admitted to TCs and NTHs in equal proportions in one state, the likelihood of dying was greater for those admitted to TCs.¹³

Nationally, lower mortality rates were found in states with formal trauma systems compared with states without formal trauma systems.¹⁵ Decreased trauma mortality after implementation of a trauma system was attributed to better performance of NTHs¹⁷ in one region, even when adjusted for severity of injury. In a state with no designated trauma system, people died at NTHs mainly after falls, while deaths at TCs were due largely to motor vehicle collisions (MVCs).¹⁹ In a comparative investigation between two neighbouring states, the overall mortality risk was similar but was significantly increased for patients who were transferred, especially in the state without TCs. In a rural state with no formal trauma system, patients with mostly severe injuries were more likely to be discharged alive from TCs.⁶ TCs were also associated with less inappropriate care and fewer preventable deaths than NTHs.¹⁸

An overview of these studies reveals that mortality risk at TCs was associated with lack of insurance, younger age and being transferred to another hospital. Mortality rates were lower for most severe cases in TCs, while higher mortality was found among very elderly patients experiencing falls in NTHs.

This study emphasizes the different characteristics of trauma fatalities in TCs compared with NTHs, and among different levels of TC in Texas, and explores the risk factors of dying at TCs and NTHs for different types of patient statewide and in rural compared with urban areas.