Methodology and Analytic Rationale for the American College of Surgeons Trauma Quality Improvement Program

The effect of the proportion of low-titer O whole blood for resuscitation in pediatric trauma patients on 6-, 12- and 24-hour survival

INTRODUCTION

Hemorrhage is a leading cause of death in pediatric patients. Accumulating data suggest that low-titer group O whole blood (LTOWB) improves clinical outcomes in the pediatric population. We examined what ratio of LTOWB to total blood product conferred a survival benefit in transfused pediatric trauma patients.

METHODS

We retrospectively examined a cohort of injured subjects younger than 18 years from the Trauma Quality Improvement Program database who received any quantity of LTOWB and no documented prehospital cardiac arrest. We created a variable representing the volume of transfused LTOWB divided by the total volume of all transfused blood products administered within the first 4 hours of admission, that is, the proportion of LTOWB transfused. We analyzed increasing proportions of transfused LTOWB to determine whether there was an inflection point conferring increased survival.

RESULTS

From 2020 to 2022, 1,122 subjects were included in the analysis. The median (interquartile range) age was 16 (14-17) years. Firearms were the most common mechanism at 47% followed by collisions at 44%. The median composite injury severity score was 25 (16-34). Survival was 91% at 6 hours, 89% at 12 hours, and 88% at 24 hours. We noted an inflection point with improved survival at an LTOWB proportion of ≥30% of total volume of blood products received. The odds of survival at 6, 12, and 24 hours for those receiving ≥30% LTOWB was 1.85 (1.02-3.38), 2.09 (1.20-3.36), and 1.80 (1.06-3.08), and 3.55 (1.66-7.58), 3.71 (1.89-7.27), and 2.69 (1.44-5.02) when excluding those who died within 1 hour, respectively.

CONCLUSION

Among LTOWB recipients, we found that a strategy of using LTOWB comprising at least 30% of the total transfusion volume within the first 4 hours was associated with improved survival at 6, 12, and 24 hours.

LEVEL OF EVIDENCE

Therapeutic/Care Management; Level III.

A national database review of whole blood use among females of childbearing potential experiencing traumatic hemorrhage

INTRODUCTION

The use of low-titer O whole blood (LTOWB) for traumatic hemorrhage is growing. Most LTOWB for use in adults is RhD-positive, which presents potential risks to females of childbearing potential (FCP); however, data on practice patterns are lacking. We sought to assess the use of LTOWB among FCPs compared to similarly aged males in facilities with documented LTOWB capabilities.

METHODS

We compared FCP (females 15-50 years of age) to similarly aged males (or sex unclassified/undocumented) who were included in the Trauma Quality Improvement Program database from 2020 to 2022. This database records transfusion volumes administered within the first 4 h after admission and patient demographics. We compared LTOWB use among FCPs versus similarly aged males using descriptive, inferential, and multivariable statistics.

RESULTS

There were 79,298 that met inclusion for this analysis. There were 16,823 (21%) FCPs, of whom, 2759/16,823 (16%) received any volume of LTOWB compared to 16,310/62,475 (26%) of the males. Furthermore, among LTOWB recipients, the median (interquartile range) volume administered to FCPs was 1162 mL (500-1000) compared to 1352 mL (500-1000, p = .003) for males. In our multivariable logistic regression analysis, males had a higher odds for the receipt of LTOWB compared to FCPs (odds ratio 1.76, 95% confidence interval 1.68-1.84) after adjusting for age, mechanism of injury, and composite injury severity score. These findings persisted on sensitivity testing.

CONCLUSIONS

Males were more likely than FCPs to receive LTOWB during trauma resuscitation in unadjusted and adjusted analyses. The reasons for such differences require elucidation in future prospective studies.

Predictive ability of frailty scores in surgically managed patients with traumatic spinal injuries: a TQIP analysis

PURPOSE

Frailty has gained recognition as a crucial determinant of patient outcomes following traumatic spinal injury (TSI), particularly due to its increasing incidence in elderly populations. The aim of the current investigation was therefore to compare the ability of several frailty scores to predict adverse outcomes in surgically managed isolated TSI patients without spinal cord injury.

METHODS

All adult patients (18 years or older) who suffered an isolated TSI due to blunt trauma, and required surgical management, were extracted from the 2013-2021 Trauma Quality Improvement Program database. The ability of the Orthopedic Frailty Score (OFS), the Hospital Frailty Risk Score (HFRS), the 11-factor (11-mFI) and 5-factor (5-mFI) modified frailty index, as well as the Johns Hopkins Frailty Indicator to predict adverse outcomes was compared based on the area under the receiver-operating characteristic curve (AUC). Subgroup analyses were also performed on patients who were ≥ 65 years old and those who were injured due to a ground-level fall (GLF).

RESULTS

A total of 39,449 patients were selected from the TQIP database. The 5-mFI and 11-mFI outperformed all other frailty scores when predicting in-hospital mortality (5-mFI AUC: 0.73) (11-mFI AUC: 0.73), any complication (5-mFI AUC: 0.65) (11-mFI AUC: 0.65), and FTR (5-mFI AUC: 0.75) (11-mFI AUC: 0.75). Among the 14,257 geriatric patients, however, the OFS demonstrated the highest predictive ability for in-hospital mortality (AUC: 0.65). The OFS (AUC: 0.64) also performed on the same level as both the 5-mFI (AUC: 0.63) and the 11-mFI (AUC: 0.63) when predicting FTR in this population. Among the 9616 patients who were injured due to a GLF, the OFS performed on par with the 5-mFI and 11-mFI when predicting in-hospital mortality and FTR.

CONCLUSION

Simpler scores like the 5-factor modified Frailty Index and Orthopedic Frailty Score outperform or perform on par with more complicated frailty scores when predicting mortality, complications, and failure-to-rescue in surgically managed isolated traumatic spinal injury patients without spinal cord injury, particularly among geriatric patients and those injured in a GLF.

The Effect of Environmental Hypothermia on Survival in Isolated Blunt Traumatic Brain Injury

BACKGROUND

Environmental hypothermia increases mortality in patients with major trauma; however, the impact of exposure hypothermia on outcomes in isolated traumatic brain injury (TBI) is underexplored in literature. The aim of this study is to determine the relationship between environmental hypothermia and survival in patients with isolated blunt TBI.

METHODS

We analyzed data from the Trauma Quality Improvement Program database. We included patients who were aged ≥15 years, had an abbreviated injury scale ≥1 for the head/neck body region, an arrival Glasgow Coma Scale of <14, an abbreviated injury scale of 0 for all other body regions, and a blunt mechanism. We defined hypothermia as <35°C.

RESULTS

From 2020 to 2022, there were 16,697 patient encounters that met inclusion for this analysis. There were 670 (4%) patient encounters that met our definition of hypothermia. Hypothermic patients had lower unadjusted survival at 24 hours (79% vs. 92%) and throughout their hospital stay (47% vs. 77%, all P < 0.001). In our multivariable logistic regression model, after adjusting for age, sex, arrival Glasgow Coma Scale, arrival shock index, mechanism of injury, and imaging findings, hypothermia was associated with lower survival at 24 hours (odds ratio: 0.59; 0.48-0.74) and lower total in-hospital survival (odds ratio: 0.44; 0.36-0.53).

CONCLUSIONS

Environmental hypothermia is associated with increased mortality at 24 hours and at hospital discharge in patients with isolated blunt TBI. Further investigation is needed to identify optimal treatment strategies for TBI patients with hypothermia and to determine whether hypothermia prevention decreases mortality.

Transfusion quantities associated with 24-h mortality in trauma patients

INTRODUCTION

Data on the correlation between transfusion volumes and trauma mortality are limited. The association between the total number of red blood cell (RBC) and low titer group O whole blood (LTOWB) units, as well as the total volume of all transfused products that were administered up to 4-h after admission and 24-h mortality was determined.

METHODS

The Trauma Quality Improvement Program (TQIP) datasets from 2020 to 2022 were reviewed to identify patients aged ≥15 who received any volume of blood products. Receiver operating characteristic (ROC) were constructed along with the calculated area under the ROC curve (AUROC) to determine the association between the quantity of transfusion and 24-h mortality.

RESULTS

There were 144,379 encounters that met inclusion, with 22,467 patients who died within the first 24 h. There was a 90% probability of 24-h mortality following the transfusion of 56 RBC/LTOWB units (AUROC 0.673), with the 90% specificity, Youden's index, and 90% sensitivity surrounding this probability occurring after the transfusion of 8, 4, and 2 units, respectively. In terms of the volume of transfusion, there was a 90% probability of 24-h mortality following the transfusion of 36,000 mL of all blood products combined (AUROC 0.662), with the 90% specificity, Youden's index, and 90% sensitivity surrounding this probability occurring after the transfusion of 4400, 2000, and 500 mL, respectively.

CONCLUSIONS

Both the total number of RBC and LTOWB units transfused and the total volume of all blood products transfused demonstrated poor predictive association with the risk of 24-h mortality in the civilian trauma population.

Risk of Severe Sepsis After Blood Product Administration for Traumatic Hemorrhage: A Trauma Quality Improvement Program Study

INTRODUCTION

Transfusion of whole blood (WB) for traumatic hemorrhage has generated renewed interest in civilian trauma based on military experience. The association between blood products and severe sepsis remains unknown. We sought to determine which blood products were associated with the development of severe sepsis.

METHODS

We utilized the TQIP database from 2020 to 2021. We included patients ≥15 ys of age who received at least one blood product and survived at least 24 hs. Severe sepsis is a standardized core quality measure for all reporting centers and defined as sepsis with organ dysfunction. We used descriptive, inferential, and multivariable logistic regression methods to test for associations and adjust for confounders.

RESULTS

There were 83,924 patients included, of whom 1471 met criteria for severe sepsis. Patients with severe sepsis tended to be older (47 versus 42, P < 0.001), male (79% versus 74%, P < 0.001), have a higher injury severity score (29 versus 19, P < 0.001), higher proportion of serious injuries to the thorax (65% versus 47%, P < 0.001), abdomen (54% versus 32%, P < 0.001), and extremities (45% versus 32%, P < 0.001). Severe sepsis patients received more packed red cells, WB, platelets, cryoprecipitate, and plasma. When adjusting for age, sex, mechanism of injury, and injury severity score, WB was positively associated with severe sepsis (unit odds ratio 1.04, 95% confidence interval 1.01-1.07).

CONCLUSIONS

Within this dataset, we found a 4% increased odds of sepsis with each unit of WB received among civilian trauma patients. The effects of blood product administration on immune system function remain unclear. High-quality, prospective explanatory studies are needed to better understand this relationship.

An Analysis of the Association of Whole Blood Transfusion With the Development of Acute Respiratory Distress Syndrome

OBJECTIVES

To determine the association of whole blood and other blood products (components, prothrombin complex concentrate, and fibrinogen concentrate) with the development of acute respiratory distress syndrome (ARDS) among blood recipients.

DESIGN

Retrospective cohort study.

SETTING

American College of Surgeons Trauma Quality Improvement Program (TQIP) database between 2020 and 2021.

PATIENTS

Patients 15 years old or older in the TQIP database between 2020 and 2022 who received at least one blood product.

INTERVENTIONS

We compared characteristics and blood product administration between patients who developed ARDS versus those who did not.

MEASUREMENTS AND MAIN RESULTS

There were 134,863 that met inclusion for this analysis. Within the included population, 1% (1927) was diagnosed with ARDS. The no ARDS group had a lower portion of serious injuries to the head/neck (31% vs. 46%), thorax (51% vs. 78%), abdomen (34% vs. 48%), and extremities (37% vs. 47%). The median composite Injury Severity Score was 21 (11-30) in the no ARDS group vs. 30 (22-41) in the ARDS group. Unadjusted survival of discharge was 74% in the no ARDS group vs. 61% in the ARDS group. In our multivariable model, we found that whole blood (unit odds ratio [uOR], 1.05; 95% CI, 1.02-1.07), male sex (odds ratio, 1.44; 95% CI, 1.28-1.63), arrival shock index (uOR, 1.03; 95% CI, 1.01-1.06), and composite Injury Severity Score (uOR, 1.03; 95% CI, 1.03-1.04) were associated with the development of ARDS. These persisted on sensitivity testing.

CONCLUSIONS

We found an association between whole blood and the development of ARDS among trauma patients who received blood transfusions. Contrary to previous studies, we found no association between ARDS and fresh frozen plasma administration. The literature would benefit from further investigation via prospective study designs.

Venous thromboembolic events associated with blood product administration in an era of whole blood use

BACKGROUND

The risks associated with blood product administration and venous thromboembolic events remains unclear. We sought to determine which blood products were associated with the development of deep vein thrombosis (DVT) and pulmonary embolism (PE).

METHODS

We analyzed data from patients ≥18 years of age in the Trauma Quality Improvement Program (TQIP) database that received ≥1 blood product and survived ≥24 ​h.

RESULTS

There were 42,399 that met inclusion, of whom, 2086 had at least one VTE event. In our multivariable logistic regression model, we found that WB had a unit odds ratio (uOR) of 1.05 (95 ​% CI 1.02-1.08) for DVT and 1.08 (1.05-1.12) for PE. Compared to WB, platelets had a higher uOR for DVT of 1.09 (1.04-1.13) but similar uOR for PE of 1.08 (1.03-1.14).

CONCLUSIONS

We found an association of both DVT and PE with early whole blood and platelets.

An assessment of nationwide trends in emergency department (ED) resuscitative endovascular balloon occlusion of the aorta (REBOA) use - A trauma quality improvement program registry analysis

BACKGROUND

Use of resuscitative endovascular balloon occlusion of the aorta (REBOA) for temporary hemorrhage control in severe non-compressible torso trauma remains controversial, with limited data on patient selection and outcomes. This study aims to analyze the nationwide trends of its use in the emergency department (EDs).

METHODS

A retrospective analysis of the American College of Surgeons Trauma Quality Improvement Program (ACS-TQIP) from 2017 to 2022 was performed, focusing on REBOA placements in EDs.

RESULTS

The analysis included 3398 REBOA procedures. Majority patients were male (76 ​%) with a median age of 40 years (27-58) and injury severity score of 20 (20-41). The most common mechanism was collision (64 ​%), with emergency surgeries most frequently performed for pelvic trauma (14 ​%). Level 1 trauma centers performed 82 ​% of these procedures, with consistent low annual utilization (<200 facilities). Survival rates were 85 ​% at 1-h post-placement, decreasing significantly to 42 ​% by discharge.

CONCLUSIONS

REBOA usage in remains limited but steady, primarily occurring at level 1 trauma center EDs. While short-term survival rates are favorable, they drop significantly by the time of discharge.

Characterising practice patterns of human derived, lyophilized coagulation concentrates within the trauma quality improvement program registry

OBJECTIVES

We seek to describe the current practice pattern use of prothrombin complex concentrate (PCC) and fibrinogen concentrate (FC) in trauma patients.

BACKGROUND

Trauma-induced coagulopathy (TIC) and endotheliopathy of trauma (EOT) contribute significantly to mortality from traumatic haemorrhage. FC, and 4-factor PCC are potential treatments for EOT and TIC, respectively.

MATERIALS AND METHODS

We obtained data from the Trauma Quality Improvement Program (TQIP) registry and identified patients who received either PCC or FC using procedural codes. We used descriptive statistics to characterise practice patterns of these products.

RESULTS

There were 6 714 002 total encounters within the TQIP from 2017 to 2022, of which 10 589 received PCC and 3009 received FC. Of the recipients, there were 35 that received both products. There were 44 that received both. The median age of PCC recipients was 77 (69-84) with 19 patients <15 years of age with the youngest being 2 years of age. There was a general upward trend in the number of facilities with documented use of PCC: 155/744, 168/766, 189/764, 206/780, 234/795, and 235/816, respectively. The median age of FC recipients was 57 (32-75) with 48 patients <15 years of age with the youngest being 1 year of age. There was a minor downward trend in the number of facilities that had documented use of FC: 55, 44, 39, 32, 38 and 40.

CONCLUSIONS

The administration of PCC and FC remains uncommon, although there appears to be an upward trend of PCC use. Most PCC use appeared to be for anticoagulation reversal in the setting of head trauma. Data guiding the use of these products are necessary as these products become more recognised as adjuncts to traumatic haemorrhage control.

Race and Socio-Economic Status Impact Withdrawal of Treatment in Young Traumatic Brain Injury

CONTEXT

Withdrawal of life-sustaining therapies (WDLST) in young individuals with traumatic brain injury (TBI) is an overwhelming situation often made more stressful by socioeconomic factors that shape health outcomes. Identifying these factors is crucial to developing equitable and goal-concordant care for patients and families.

OBJECTIVES

We aimed to identify predictors of WDLST in young patients with 1-TBI. We hypothesized uninsured payment method, race, and co-morbid status are associated with WDLST.

METHODS

We queried the 2021 Trauma Quality Improvement Program database for patients <45 years with TBI. Patients with WDLST were compared to patients without WDLST. Multivariable logistic regression (MLR) was performed.

RESULTS

61,115 patients were included, of whom 2,487 (4.1%) underwent WDLST. Patients in the WDLST cohort were older (29 vs 27, P<0.001), more likely to suffer from a penetrating mechanism (29% vs 11%, P<.0001), and have uninsured (22% vs 18%) or other payment method (5% vs 3%) when compared to the non-WDLST cohort. MLR identified age (AOR:1.019, 95% CI 1.014-1.024, P<.0001), non-Hispanic ethnicity (AOR:1.590, 95% CI 1.373-1.841, P<.0001), penetrating mechanism (AOR:3.075, 95% CI 2.727-3.467, P<.0001), systolic blood pressure (AOR: 0.992, 95% CI 0.990-0.993, P<0.0001), advanced directive (AOR:4.987, 95% CI 2.823-8.812, P<.0001), cirrhosis (AOR:3.854, 95% CI 2.641-5.625, P<.0001), disseminated cancer (AOR:6.595, 95% CI 2.370-18.357, P=0.0003), and interfacility transfer (AOR:1.457, 95% CI 1.295-1.640, P<0.0001) as factors associated with WDLST. Black patients were less likely to undergo WDLST when compared to white patients (AOR:0.687, 95% CI 0.603-0.782, P<.0001).

CONCLUSION

The decision for WDLST in young patients with severe TBI may be influenced by cultural and socioeconomic factors in addition to clinical considerations.

Are trauma centers penalized for improved prehospital resuscitation?: The effect of prehospital transfusion on arrival vitals and predicted mortality

BACKGROUND

Prediction models for survival in trauma rely on arrival vital signs to generate survival probabilities. Hospitals are benchmarked on expected and observed outcomes. Prehospital blood (PB) transfusion has been shown to improve mortality, which may affect survival prediction modeling. We hypothesize that the use of PB increases the predicted survival derived from probability models compared with non-blood-based resuscitation.

METHODS

All adult trauma patients presenting to a level 1 trauma center requiring emergency release blood transfusion from January 2017 to December 2021 were reviewed. Patients were grouped into those receiving PB or those who did not (no PB). Prehospital Trauma and Injury Severity Score (TRISS) and shock index were compared with those at presentation to hospital. Univariate and multivariate regressions were performed to identify factors associated with changes in survival probability at presentation.

RESULTS

In total, 2117 patients were reviewed (PB, 1,011; no PB, 1,106). Patients receiving PB were younger (35 vs. 40 years, p < 0.001), more likely to have blunt mechanism (71% vs. 65%, p = 0.002), and more severely injured (Injury Severity Score, 27 vs. 25; p < 0.001) and had higher rates of prehospital hypotension (44% vs. 19%, p < 0.001) and shock index (1.10 vs. 0.87, p < 0.001). Upon arrival, PB patients had lower rates of ED hypotension (34% vs. 39%, p = 0.01), and significant improvements in arrival TRISS scores (+0.09 vs. -0.02, p < 0.001) and shock index (+0.10 vs. -0.07, p < 0.001) compared with prehospital. On multivariate analysis, PB was associated with a threefold increase in unexpected survivors (odds ratio, 3.28; 95% confidence interval, 2.23-4.60).

CONCLUSION

The use of PB was associated with improved probability of survival and an increase in unexpected survivors. Applying TRISS and shock index at hospital arrival does not account for en route hemostatic resuscitation, causing patients to arrive with improved vitals despite severity of injury. Caution should be used when implementing survival probability calculations using arrival vitals in centers with prehospital transfusion capability.

LEVEL OF EVIDENCE

Therapeutic/Care Management; Level IV.

Socioeconomic disadvantage is a leading variable in risk score for major amputation following emergent infrainguinal arterial bypass surgery

OBJECTIVE

The purpose of this study was to identify patients at particularly high risk for major amputation after emergent infrainguinal bypass to help tailor postoperative and long-term patient management.

METHODS

In the Vascular Quality Initiative, we identified 2126 patients who underwent emergent infrainguinal artery bypass. Two primary outcomes were investigated: major ipsilateral amputation above the ankle level during the index hospitalization and major amputation above the ankle at any time after emergent infrainguinal bypass surgery (perioperative and postdischarge combined). Binary logistic regression analysis was performed for each outcome using variables that achieved a univariable P value of ≤.10. We then determined which variables have a multivariable association for the outcomes as defined by a regression P value of ≤.05. A risk score was then created for the outcome of amputation after emergent infrainguinal bypass using weighted beta-coefficient. Variables with a multivariable P value of ≤.05 were included in the risk score and weighted based on their respective regression beta-coefficient in a point scale.

RESULTS

Overall, 17.1% of patients (368/2126) underwent major amputation at some point in follow-up after emergent infrainguinal artery bypass. The mean follow-up duration on the amputation variable was 261 days with the end point being time of amputation or time of last follow-up data on the amputation variable. Variables with a significant multivariable association (P < .05) with major amputation at any point after emergent infrainguinal arterial bypass were home status in top 10% (most deprived) of Area Deprivation Index, prior infrainguinal ipsilateral arterial bypass, prior ipsilateral endovascular arterial intervention, prosthetic bypass conduit, postoperative skin/soft tissue infection, and postoperative need to revise or thrombectomize bypass. Pertinent negatives on multivariable analysis included all baseline comorbidities, insurance status, race, and gender. There is steep progression in amputation rate ranging from 5% at scores of 0 and 1 to >60% for scores in of >10. Area under the curve analysis revealed a value of 0.706.

CONCLUSIONS

Patients living in the most disadvantaged socioeconomic neighborhoods have an increased risk of amputation after emergent infrainguinal arterial bypass independent of baseline comorbidities and perioperative events. Baseline comorbidities are not impactful regarding amputation rates after emergent infrainguinal bypass surgery. The need for bypass revision or thrombectomy during the index hospitalization is the most impactful factor toward amputation after emergency bypass. A risk score with quality accuracy has been developed to help identify patients at particularly high likelihood of limb loss, which may aid in counseling regarding heightened vigilance in postoperative and long-term follow-up care.

An analysis of the effect of low titer O whole blood (LTOWB) proportions for resuscitation after trauma on 6-hour and 24-hour survival

INTRODUCTION

Hemorrhage is a leading cause of death. Blood products are used for the treatment of hemorrhagic shock. The use of low titer group O whole blood (LTOWB) has become more common.

METHODS

Data from patients ≥15 years of age in the Trauma Quality Improvement Program (TQIP) database that received ≥10 units of packed red cells and/or LTOWB within the first 4-h of hospital arrival were included. The proportion of LTWOB of total blood products administered was correlated to 6- and 24-h mortality.

RESULTS

12,763 met inclusion, 3827 (30 %) received LTOWB. On multivariable logistic regression (MVLR), there was no difference in survival at 6 h with a LTOWB. When assessing 24-h survival, there was improved survival with LTOWB ≥10 % (OR 1.18, 1.08-1.28).

CONCLUSIONS

In this analysis of TQIP data, patients receiving ≥10 units of PRBC or LTOWB, we found that higher proportions of LTOWB transfusion relative to the total volume of blood products transfused during the first 4 h were associated with improved 24-h, but not 6-h survival.

Characterizing emergency department surgical airway placement in the setting of trauma

INTRODUCTION

Airway management is a key intervention during the resuscitation of critically ill trauma patients. Emergency surgical airway (ESA) placement is taught as a backup option when endotracheal intubation (ETI) fails. We sought to (1) describe the incidence of the emergency department (ED) ESA, (2) compare ESA versus ETI-only recipients, and (3) determine which factors were associated with receipt of an ESA.

METHODS

We searched within the Trauma Quality Improvement Program datasets from 2017 to 2022 for all emergency department surgical airway placement and/or endotracheal intubations recipients. We compared ESA versus ETI-only recipients.

RESULTS

From 2017 to 2022, there were 6,477,759 within the datasets, of which 238,128 met inclusion for this analysis. Within that, there were 236,292 ETIs, 2264 ESAs, with 428 (<1 %) having documentation of both. Of the ESAs performed, there were 82 documented in children <15 years of age with the youngest being 1 year of age. The ETI-only group had a lower proportion serious injuries to the head/neck (52 % versus 59 %), face (2 % versus 8 %), and skin (3 % versus 6 %). However, the ETI-only group had a higher proportion of serious injuries to the abdomen (15 % versus 9 %) and the extremities (19 % versus 12 %). Survival at 24-h was higher in the ETI-only group (83 % versus 76 %) as well as survival to discharge (70 % versus 67 %). In the subanaysis of children <15 years (n = 82), 34 % occurred in the 1-4 years age group, 35 % in the 5-9 years age group, and 30 % in the 10-14 years age group. In our multivariable logistic regression analysis, serious injuries to the head/neck (odds ratio [OR] 1.37, 95 % CI 1.23-1.54), face (OR 3.41, 2.83-4.11), thorax (OR 1.19, 1.06-1.33), and skin (OR 1.53, 1.15-2.05) were all associated with receipt of cricothyrotomy. Firearm (OR 3.62, 3.18-4.12), stabbing (2.85, 2.09-3.89), and other (OR 2.85, 2.09-3.89) were associated with receipt of ESA when using collision as the reference variable.

CONCLUSIONS

ESA placement is a rarely performed procedure but frequently used as a primary airway intervention in this dataset. Penetrating mechanisms, and injuries to face were most associated with ESA placement. Our findings reinforce the need to maintain this critical airway skill for trauma management.

A multidimensional approach to identifying high-performing trauma centers across the United States

INTRODUCTION

The differentiators of centers performing at the highest level of quality and patient safety are likely both structural and cultural. We aimed to combine five indicators representing established domains of trauma quality and to identify and describe the structural characteristics of consistently performing centers.

METHODS

Using American College of Surgeons Trauma Quality Improvement Program data from 2017 to 2020, we evaluated five quality measures across several care domains for adult patients in levels I and II trauma centers: (1) time to operating room for patients with abdominal gunshot wounds and shock, (2) proportion of patients receiving timely venous thromboembolism prophylaxis, (3) failure to rescue (death following a complication), (4) major hospital complications, and (5) mortality. Overall performance was summarized as a composite score incorporating all measures. Centers were ranked from highest to lowest performer. Principal component analysis showed the influence of each indicator on overall performance and supported the composite score approach.

RESULTS

We identified 272 levels I and II centers, with 28 and 27 centers in the top and bottom 10%, respectively. Patients treated in high-performing centers had significant lower rates of death major complications and failure to rescue, compared with low-performing centers ( p < 0.001). The median time to operating room for gunshot wound was almost half that in high compared with low-performing centers, and rates of timely venous thromboembolism prophylaxis were over twofold greater ( p < 0.001). Top performing centers were more likely to be level I centers and cared for a higher number of severely injured patients per annum. Each indicator contributed meaningfully to the variation in scores and centers tended to perform consistently across most indicators.

CONCLUSION

The combination of multiple indicators across dimensions of quality sets a higher standard for performance evaluation and allows the discrimination of centers based on structural elements, specifically level 1 status, and trauma center volume.

LEVEL OF EVIDENCE

Therapeutic /Care Management; Level IV.

Changes in Emergency Department Pediatric Readiness and Mortality

Importance

High emergency department (ED) pediatric readiness is associated with improved survival, but the impact of changes to ED readiness is unknown.

Objective

To evaluate the association of changes in ED pediatric readiness at US trauma centers between 2013 and 2021 with pediatric mortality.

Design, Setting, and Participants

This retrospective cohort study was performed from January 1, 2012, through December 31, 2021, at EDs of trauma centers in 48 states and the District of Columbia. Participants included injured children younger than 18 years with admission or injury-related death at a participating trauma center, including transfers to other trauma centers. Data analysis was performed from May 2023 to January 2024.

Exposure

Change in ED pediatric readiness, measured using the weighted Pediatric Readiness Score (wPRS, range 0-100, with higher scores denoting greater readiness) from national assessments in 2013 and 2021. Change groups included high-high (wPRS ≥93 on both assessments), low-high (wPRS <93 in 2013 and wPRS ≥93 in 2021), high-low (wPRS ≥93 in 2013 and wPRS <93 in 2021), and low-low (wPRS <93 on both assessments).

Main Outcomes and Measures

The primary outcome was lives saved vs lost, according to ED and in-hospital mortality. The risk-adjusted association between changes in ED readiness and mortality was evaluated using a hierarchical, mixed-effects logistic regression model based on a standardized risk-adjustment model for trauma, with a random slope-random intercept to account for clustering by the initial ED.

Results

The primary sample included 467 932 children (300 024 boys [64.1%]; median [IQR] age, 10 [4 to 15] years; median [IQR] Injury Severity Score, 4 [4 to 15]) at 417 trauma centers. Observed mortality by ED readiness change group was 3838 deaths of 144 136 children (2.7%) in the low-low ED group, 1804 deaths of 103 767 children (1.7%) in the high-low ED group, 1288 deaths of 64 544 children (2.0%) in the low-high ED group, and 2614 deaths of 155 485 children (1.7%) in the high-high ED group. After risk adjustment, high-readiness EDs (persistent or change to) had 643 additional lives saved (95% CI, -328 to 1599 additional lives saved). Low-readiness EDs (persistent or change to) had 729 additional preventable deaths (95% CI, -373 to 1831 preventable deaths). Secondary analysis suggested that a threshold of wPRS 90 or higher may optimize the number of lives saved. Among 716 trauma centers that took both assessments, the median (IQR) wPRS decreased from 81 (63 to 94) in 2013 to 77 (64 to 93) in 2021 because of reductions in care coordination and quality improvement.

Conclusions and Relevance

Although the findings of this study of injured children in US trauma centers were not statistically significant, they suggest that trauma centers should increase their level of ED pediatric readiness to reduce mortality and increase the number of pediatric lives saved after injury.

A survey of low titer O whole blood use within the trauma quality improvement program registry

INTRODUCTION

The use of low titer O whole blood (LTOWB) has expanded although it remains unclear how many civilian trauma centers are using LTOWB.

METHODS

We analyzed data on civilian LTOWB recipients in the American College of Surgeons Trauma Quality Improvement Program (TQIP) database 2020-2021. Unique facility keys were used to determine the number of centers that used LTOWB in that period.

RESULTS

A total of 16,603 patients received LTOWB in the TQIP database between 2020 and 2021; 6600 in 2020, and 10,003 in 2021. The total number of facilities that reported LTOWB use went from 287/779 (37%) in 2020 to 302/795 (38%) in 2021. Between 2020 and 2021, among all level 1-3 designated trauma facilities that report to TQIP LTOWB use increased at level-1 centers (118 to 129), and level-2 centers (81 to 86), but decreased in level-3 facilities (9 to 4). Among pediatric and dual pediatric-adult designated hospitals there was a decrease in the number of pediatric level-1 centers (29 to 28) capable of administering LTOWB. Among centers with either single or dual level-1 trauma center designation with adult centers, the number that administered LTOWB to injured pediatric patients also decreased from 17 to 10, respectively.

CONCLUSIONS

There was an increase in the number of facilities transfusing LTOWB between 2020 and 2021. The use of LTOWB is underutilized in children at centers that have it available. These findings inform the expansion of LTOWB use in trauma.

Complication reporting in trauma: An environmental scan and comparison of nationwide trauma registry data

BACKGROUND

To explore variability in quality measurement, this study aimed to compare abstraction and definitions of complications reported across trauma registries in Canada.

METHODS

A literature search was performed to identify active trauma registries used in Canadian hospitals. Registry characteristics, data abstraction, and reported complications and definitions based on registry data dictionaries were compared.

RESULTS

Nine registries were included, most of which were provincial-level registries (67 ​%). A total of 53 individual complications were identified. Twenty-one (40 ​%) were recorded by only one registry each whereas 5 (9 ​%) were collected by all. Of the 32 complications collected by ​> ​1 registry, 18 (56 ​%) had different definitions. Of the 18 with different definitions, 12 (67 ​%), 5 (28 ​%), and 1 (6 ​%) had 2, 3, and 4 different definitions across registries, respectively.

CONCLUSIONS

Complications reported by trauma registries are variable. Reliable benchmarking is likely challenging, and efforts to standardize complication reporting may be a valuable undertaking.

Treatment of Acute Traumatic Central Cord Syndrome: A Study of North American Trauma Centers

BACKGROUND AND OBJECTIVES

Central cord syndrome (CCS) is expected to become the most common traumatic spinal cord injury, yet its optimal management remains unclear. This study aimed to evaluate variability in nonoperative vs operative treatment for CCS between trauma centers in the American College of Surgeons Trauma Quality Improvement Program, identify patient- and hospital-level factors associated with treatment, and determine the association of treatment with outcomes.

METHODS

Adults with CCS were identified from the Trauma Quality Improvement Program database (2014-2016). Mixed-effects modeling with a random intercept for trauma centers was used to examine the adjusted association of patient- and hospital-level variables with nonoperative treatment. The random-effects output of the model assessed the risk-adjusted variability in nonoperative treatment across centers. Outlier hospitals were identified, and the median odds ratio was calculated. The adjusted effect of nonoperative treatment on mortality, morbidity, and hospital length of stay (LOS) was examined at the patient and hospital level by mixed-effects regression.

RESULTS

Three thousand, nine hundred twenty-eight patients across 255 centers were eligible; of these, 1523 (38.8%) were treated nonoperatively. Older age, noncommercial insurance (odds ratio [OR] 1.26, 95% CI 1.08-1.48, P = .004), absence of fracture (OR 0.58, 95% CI 0.49-0.68, P < .001), severe head injury (OR 1.41, 95% CI 1.09-1.82, P = .008), and comatose presentation (1.82, 95% CI 1.15-2.89, P = .011) were associated with nonoperative treatment. Twenty-eight hospitals were outliers, and the median odds ratio was 2.02. Patients receiving nonoperative treatment had shorter LOS (mean difference -4.65 days). Nonoperative treatment was associated with lesser in-hospital morbidity (OR 0.49, 95% CI 0.37-0.63, P < .001) at the patient level. There was no difference in mortality.

CONCLUSION

Operative decision-making for CCS is influenced by patient factors. There remains substantial variability between trauma centers not explained by case-mix differences. Nonoperative treatment was associated with shorter hospital LOS and lesser inpatient morbidity.

Early Vital Sign Thresholds Associated with 24-Hour Mortality among Trauma Patients: A Trauma Quality Improvement Program (TQIP) Study

BACKGROUND

Identifying patients at imminent risk of death is critical in the management of trauma patients. This study measures the vital sign thresholds associated with death among trauma patients.

METHODS

This study included data from patients ≥15 years of age in the American College of Surgeons Trauma Quality Improvement Program (TQIP) database. Patients with vital signs of zero were excluded. Documented prehospital and emergency department (ED) vital signs included systolic pressure, heart rate, respiratory rate, and calculated shock index (SI). The area under the receiver operator curves (AUROC) was used to assess the accuracy of these variables for predicting 24-hour survival. Optimal thresholds to predict mortality were identified using Youden's Index, 90% specificity, and 90% sensitivity. Additional analyses examined patients 70+ years of age.

RESULTS

There were 1,439,221 subjects in the 2019-2020 datasets that met inclusion for this analysis with <0.1% (10,270) who died within 24 hours. The optimal threshold for prehospital systolic pressure was 110, pulse rate was 110, SI was 0.9, and respiratory rate was 15. The optimal threshold for the ED systolic was 112, pulse rate was 107, SI was 0.9, and respiratory rate was 21. Among the elderly sub-analysis, the optimal threshold for prehospital systolic was 116, pulse rate was 100, SI was 0.8, and respiratory rate was 21. The optimal threshold for ED systolic was 121, pulse rate was 95, SI was 0.8, and respiratory rate was 21.

CONCLUSIONS

Systolic blood pressure (SBP) and SI offered the best predictor of mortality among trauma patients. The SBP values predictive of mortality were significantly higher than the traditional 90mmHg threshold. This dataset highlights the need for better methods to guide resuscitation as initial vital signs have limited accuracy in predicting subsequent mortality.

Benchmarking Pediatric Trauma Care in Mixed Trauma Centers: Adult Risk-Adjusted Mortality Is Not a Reliable Indicator of Pediatric Outcomes

BACKGROUND

Trauma center benchmarking has become standard practice for assessing quality. The American College of Surgeons adult trauma center verification standards do not specifically require participation in a pediatric-specific benchmarking program. Centers that treat adults and children may therefore rely solely on adult benchmarking metrics as a surrogate for pediatric quality. This study assessed discordance between adult and pediatric mortality within mixed trauma centers to determine the need to independently report pediatric-specific quality metrics.

STUDY DESIGN

A cohort of trauma centers (n = 493, including 347 adult-only, 44 pediatric-only, and 102 mixed) that participated in the American College of Surgeons TQIP in 2017 to 2018 was analyzed. Center-specific observed-to-expected mortality estimates were calculated using TQIP adult inclusion criteria for 449 centers treating adults (16 to 65 years) and using TQIP pediatric inclusion criteria for 146 centers treating children (0 to 15 years). We then correlated risk-adjusted mortality estimates for pediatric and adult patients within mixed centers and evaluated concordance of their outlier status between adults and children.

RESULTS

The cohort included 394,075 adults and 97,698 children. Unadjusted mortality was 6.1% in adults and 1.2% in children. Mortality estimates had only moderate correlation ( r = 0.41) between adult and pediatric cohorts within individual mixed centers. Mortality outlier status for adult and pediatric cohorts was discordant in 31% (32 of 102) of mixed centers (weighted Kappa statistic 0.06 [-0.11 to 0.22]), with 78% (23 of 32) of discordant centers having higher odds of mortality for children than for adults (6 centers with average adult mortality and high pediatric mortality and 17 centers with low adult mortality and average pediatric mortality, p < 0.01).

CONCLUSIONS

Adult mortality is not a reliable surrogate for pediatric mortality in mixed trauma centers. Incorporation of pediatric-specific benchmarks should be required for centers that admit children.

The VANCO Trial Findings Are Generalizable to a North American Trauma Registry

OBJECTIVES

To estimate the generalizability of treatment effects observed in the VANCO trial to a broader population of patients with tibial plateau or pilon fractures.

METHODS

Design and Setting: Clinical trial data from 36 United States trauma centers and Trauma Quality Programs registry data from more than 875 Level I-III trauma centers in the United States and Canada.Patient Selection Criteria: Patients enrolled in the VANCO trial treated with intrawound vancomycin powder from January 2015 to June 2017 and 31,924 VANCO-eligible TQP patients admitted in 2019 with tibial plateau and pilon fractures.Outcome Measure and Comparisons: Deep surgical site infection and gram-positive deep surgical site infection estimated in the TQP sample weighed by the inverse probability of trial participation.

RESULTS

The 980 patients in the VANCO trial were highly representative of 31,924 TQP VANCO-eligible patients (Tipton generalizability index 0.96). It was estimated that intrawound vancomycin powder reduced the odds of deep surgical infection by odds ratio (OR) = 0.46 (95% confidence interval [CI] 0.25-0.86) and gram-positive deep surgical infection by OR = 0.39 (95% CI, 0.18-0.84) within the TQP sample of VANCO-eligible patients. For reference, the trial average treatment effects for deep surgical infection and gram-positive deep surgical infection were OR = 0.60 (95% CI, 0.37-0.98) and OR = 0.44 (95% CI, 0.23-0.80), respectively.

CONCLUSIONS

This generalizability analysis found that the inferences of the VANCO trial generalize and might even underestimate the effects of intrawound vancomycin powder when observed in a wider population of patients with tibial plateau and pilon fractures.

LEVEL OF EVIDENCE

Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Earlier Tracheostomy Reduces Complications in Complete Cervical Spinal Cord Injury in Real-World Practice: Analysis of a Multicenter Cohort of 2001 Patients

BACKGROUND AND OBJECTIVES

It is believed that early tracheostomy in patients with traumatic cervical spinal cord injury (SCI) may lessen the risk of developing complications and reduce the duration of mechanical ventilation and critical care stay. This study aims to assess whether early tracheostomy is beneficial in patients with traumatic cervical SCI.

METHODS

We conducted a retrospective cohort study using data from the American College of Surgeons Trauma Quality Improvement Program database from 2010 to 2018. Adult patients with a diagnosis of acute complete (ASIA A) traumatic cervical SCI who underwent surgery and tracheostomy were included. Patients were stratified into those receiving early (at or before 7 days) and delayed tracheostomy. Propensity score matching was used to assess the association between delayed tracheostomy and the risk of in-hospital adverse events. Risk-adjusted variability in tracheostomy timing across trauma centers was investigated using mixed-effects regression.

RESULTS

The study included 2001 patients from 374 North American trauma centers. The median time to tracheostomy was 9.2 days (IQR: 6.1-13.1 days), with 654 patients (32.7%) undergoing early tracheostomy. After matching, the odds of a major complication were significantly lower for early tracheostomy patients (OR: .90; 95% CI: .88-.98). Patients were also significantly less likely to experience an immobility-related complication (OR: .90; 95% CI: .88-.98). Patients in the early group spent 8.2 fewer days in the critical care unit (95% CI: -10.2 to -6.61) and 6.7 fewer days ventilated (95% CI: -9.44 to -5.23). There was significant variability in tracheostomy timeliness between trauma centers with a median odds ratio of 12.2 (95% CI: 9.7-13.7), which was not explained by case-mix and hospital-level characteristics.

CONCLUSION

A 7-day threshold to implement tracheostomy seems to be associated with reduced in-hospital complications, time in the critical care unit, and time on mechanical ventilation.

Trauma Quality Improvement Program: A Retrospective Analysis from A Middle Eastern National Trauma Center

BACKGROUND

The Trauma Quality Improvement Program (American College of Surgery (ACS-TQIP)) uses the existing infrastructure of the Committee on Trauma programs and provides feedback to participating hospitals on risk-adjusted outcomes. This study aimed to analyze and compare the performance of the Level I Hamad Trauma Centre (HTC) with other TQIP participating centers by comparing TQIP aggregate database reports. The primary goal was to pinpoint the variations in adult trauma outcomes and quality measures, identify areas that need improvement, and leverage existing resources to facilitate quality improvement.

METHODS

A retrospective analysis was performed for the TQIP data from April 2019-March 2020 to April 2020-March 2021. We used the TQIP methodology, inclusion and exclusion criteria, and outcomes.

RESULTS

There were 915 patients from Fall 2020 and 884 patients from Fall 2021 that qualified for the TQIP database. The HTC patients' demographics differed from the TQIP's aggregate data; they were younger, more predominantly male, and had significantly different mechanisms of injury (MOI) with more traffic-related blunt trauma. Penetrating injuries were more severe in the other centers. During the TQIP Fall 2020 report, the HTC was a low outlier (good performer) in one cohort (all patients) and an average performer in the remaining cohorts. However, during Fall 2021, the HTC showed an improvement and was a low outlier in two cohorts (all patients and severe TBI patients). Overall, the HTC remained an average performer during the report cycles.

CONCLUSIONS

There was an improvement over time in the risk-adjusted mortality, which reflects the continuous and demanding effort put together by the trauma team. The ACS-TQIP for the external benchmarking of quality improvement could be a contributor to better monitored patient care. Evaluating the TQIP data with emphases on appropriate methodologies, quality measurements, corrective measures, and accurate reporting is warranted.

Association of Transport Time, Proximity, and Emergency Department Pediatric Readiness With Pediatric Survival at US Trauma Centers

Importance

Emergency department (ED) pediatric readiness is associated with improved survival among children. However, the association between geographic access to high-readiness EDs in US trauma centers and mortality is unclear.

Objective

To evaluate the association between the proximity of injury location to receiving trauma centers, including the level of ED pediatric readiness, and mortality among injured children.

Design, Setting, and Participants

This retrospective cohort study used a standardized risk-adjustment model to evaluate the association between trauma center proximity, ED pediatric readiness, and in-hospital survival. There were 765 trauma centers (level I-V, adult and pediatric) that contributed data to the National Trauma Data Bank (January 1, 2012, through December 31, 2017) and completed the 2013 National Pediatric Readiness Assessment (conducted from January 1 through August 31, 2013). The study comprised children aged younger than 18 years who were transported by ground to the included trauma centers. Data analysis was performed between January 1 and March 31, 2022.

Exposures

Trauma center proximity within 30 minutes by ground transport and ED pediatric readiness, as measured by weighted pediatric readiness score (wPRS; range, 0-100; quartiles 1 [low readiness] to 4 [high readiness]).

Main Outcomes and Measures

In-hospital mortality. We used a patient-level mixed-effects logistic regression model to evaluate the association of transport time, proximity, and ED pediatric readiness on mortality.

Results

This study included 212 689 injured children seen at 765 trauma centers. The median patient age was 10 (IQR, 4-15) years, 136 538 (64.2%) were male, and 127 885 (60.1%) were White. A total of 4156 children (2.0%) died during their hospital stay. The median wPRS at these hospitals was 79.1 (IQR, 62.9-92.7). A total of 105 871 children (49.8%) were transported to trauma centers with high-readiness EDs (wPRS quartile 4) and another 36 330 children (33.7%) were injured within 30 minutes of a quartile 4 ED. After adjustment for confounders, proximity, and transport time, high ED pediatric readiness was associated with lower mortality (highest-readiness vs lowest-readiness EDs by wPRS quartiles: adjusted odds ratio, 0.65 [95% CI, 0.47-0.89]). The survival benefit of high-readiness EDs persisted for transport times up to 45 minutes. The findings suggest that matching children to trauma centers with high-readiness EDs within 30 minutes of the injury location may have potentially saved 468 lives (95% CI, 460-476 lives), but increasing all trauma centers to high ED pediatric readiness may have potentially saved 1655 lives (95% CI, 1647-1664 lives).

Conclusions and Relevance

These findings suggest that trauma centers with high ED pediatric readiness had lower mortality after considering transport time and proximity. Improving ED pediatric readiness among all trauma centers, rather than selective transport to trauma centers with high ED readiness, had the largest association with pediatric survival. Thus, increased pediatric readiness at all US trauma centers may substantially improve patient outcomes after trauma.

Emergency Department Pediatric Readiness and Disparities in Mortality Based on Race and Ethnicity

Importance

Presentation to emergency departments (EDs) with high levels of pediatric readiness is associated with improved pediatric survival. However, it is unclear whether children of all races and ethnicities benefit equitably from increased levels of such readiness.

Objective

To evaluate the association of ED pediatric readiness with in-hospital mortality among children of different races and ethnicities with traumatic injuries or acute medical emergencies.

Design, Setting, and Participants

This cohort study of children requiring emergency care in 586 EDs across 11 states was conducted from January 1, 2012, through December 31, 2017. Eligible participants included children younger than 18 years who were hospitalized for an acute medical emergency or traumatic injury. Data analysis was conducted between November 2022 and April 2023.

Exposure

Hospitalization for acute medical emergency or traumatic injury.

Main Outcomes and Measures

The primary outcome was in-hospital mortality. ED pediatric readiness was measured through the weighted Pediatric Readiness Score (wPRS) from the 2013 National Pediatric Readiness Project assessment and categorized by quartile. Multivariable, hierarchical, mixed-effects logistic regression was used to evaluate the association of race and ethnicity with in-hospital mortality.

Results

The cohort included 633 536 children (median [IQR] age 4 [0-12] years]). There were 557 537 children (98 504 Black [17.7%], 167 838 Hispanic [30.1%], 311 157 White [55.8%], and 147 876 children of other races or ethnicities [26.5%]) who were hospitalized for acute medical emergencies, of whom 5158 (0.9%) died; 75 999 children (12 727 Black [16.7%], 21 604 Hispanic [28.4%], 44 203 White [58.2%]; and 21 609 of other races and ethnicities [27.7%]) were hospitalized for traumatic injuries, of whom 1339 (1.8%) died. Adjusted mortality of Black children with acute medical emergencies was significantly greater than that of Hispanic children, White children, and of children of other races and ethnicities (odds ratio [OR], 1.69; 95% CI, 1.59-1.79) across all quartile levels of ED pediatric readiness; but there were no racial or ethnic disparities in mortality when comparing Black children with traumatic injuries with Hispanic children, White children, and children of other races and ethnicities with traumatic injuries (OR 1.01; 95% CI, 0.89-1.15). When compared with hospitals in the lowest quartile of ED pediatric readiness, children who were treated at hospitals in the highest quartile had significantly lower mortality in both the acute medical emergency cohort (OR 0.24; 95% CI, 0.16-0.36) and traumatic injury cohort (OR, 0.39; 95% CI, 0.25-0.61). The greatest survival advantage associated with high pediatric readiness was experienced for Black children in the acute medical emergency cohort.

Conclusions and Relevance

In this study, racial and ethnic disparities in mortality existed among children treated for acute medical emergencies but not traumatic injuries. Increased ED pediatric readiness was associated with reduced disparities; it was estimated that increasing the ED pediatric readiness levels of hospitals in the 3 lowest quartiles would result in an estimated 3-fold reduction in disparity for pediatric mortality. However, increased pediatric readiness did not eliminate disparities, indicating that organizations and initiatives dedicated to increasing ED pediatric readiness should consider formal integration of health equity into efforts to improve pediatric emergency care.

Using medicare claims to estimate risk-adjusted performance of Pennsylvania trauma centers

Trauma centers use registry data to benchmark performance using a standardized risk adjustment model. Our objective was to utilize national claims to develop a risk adjustment model applicable across all hospitals, regardless of designation or registry participation. Patients from 2013-14 Pennsylvania Trauma Outcomes Study (PTOS) registry data were probabilistically matched to Medicare claims using demographic and injury characteristics. Pairwise comparisons established facility linkages and matching was then repeated within facilities to link records. Registry models were estimated using GLM and compared with five claims-based LASSO models: demographics, clinical characteristics, diagnosis codes, procedures codes, and combined demographics/clinical characteristics. Area under the curve and correlation with registry model probability of death were calculated for each linked and out-of-sample cohort. From 29 facilities, a cohort comprising 16,418 patients were linked between datasets. Patients were similarly distributed: median age 82 (PTOS IQR: 74-87 vs. Medicare IQR: 75-88); non-white 6.2% (PTOS) vs. 5.8% (Medicare). The registry model AUC was 0.86 (0.84-0.87). Diagnosis and procedure codes models performed poorest. The demographics/clinical characteristics model achieved an AUC = 0.84 (0.83-0.86) and Spearman = 0.62 with registry data. Claims data can be leveraged to create models that accurately measure the performance of hospitals that treat trauma patients.

Assessment of the incremental prognostic value from the modified frailty index-5 in complete traumatic cervical spinal cord injury

Frailty, as measured by the modified frailty index-5 (mFI-5), and older age are associated with increased mortality in the setting of spinal cord injury (SCI). However, there is limited evidence demonstrating an incremental prognostic value derived from patient mFI-5. We conducted a retrospective cohort study to evaluate in-hospital mortality among adult complete cervical SCI patients at participating centers of the Trauma Quality Improvement Program from 2010 to 2018. Logistic regression was used to model in-hospital mortality, and the area under the receiver operating characteristic curve (AUROC) of regression models with age, mFI-5, or age with mFI-5 was used to compare the prognostic value of each model. 4733 patients were eligible. We found that both age (80 y versus 60 y: OR 3.59 95% CI [2.82 4.56], P < 0.001) and mFI-5 (score ≥ 2 versus < 2: OR 1.53 95% CI [1.19 1.97], P < 0.001) had statistically significant associations with in-hospital mortality. There was no significant difference in the AUROC of a model including age and mFI-5 when compared to a model including age without mFI-5 (95% CI Δ AUROC [- 8.72 × 10^-4 0.82], P = 0.199). Both models were superior to a model including mFI-5 without age (95% CI Δ AUROC [0.06 0.09], P < 0.001). Our findings suggest that mFI-5 provides minimal incremental prognostic value over age with respect to in-hospital mortality for patients complete cervical SCI.

Impact of individual components of emergency department pediatric readiness on pediatric mortality in US trauma centers

BACKGROUND

Injured children initially treated at trauma centers with high emergency department (ED) pediatric readiness have improved survival. Centers with limited resources may not be able to address all pediatric readiness deficiencies, and there currently is no evidence-based guidance for prioritizing different components of readiness. The objective of this study was to identify individual components of ED pediatric readiness associated with better-than-expected survival in US trauma centers to aid in the allocation of resources targeted at improving pediatric readiness.

METHODS

This cohort study of US trauma centers used the National Trauma Data Bank (2012-2017) matched to the 2013 National Pediatric Readiness Project assessment. Adult and pediatric centers treating at least 50 injured children (younger than 18 years) and recording at least one death during the 6-year study period were included. Using a standardized risk-adjustment model for trauma, we calculated the observed-to-expected mortality ratio for each trauma center. We used bivariate analyses and multivariable linear regression to assess for associations between individual components of ED pediatric readiness and better-than-expected survival.

RESULTS

Among 555 trauma centers, the observed-to-expected mortality ratios ranged from 0.07 to 4.17 (interquartile range, 0.93-1.14). Unadjusted analyses of 23 components of ED pediatric readiness showed that trauma centers with better-than-expected survival were more likely to have a validated pediatric triage tool, comprehensive quality improvement processes, a pediatric-specific disaster plan, and critical airway and resuscitation equipment (all p < 0.03). The multivariable analysis demonstrated that trauma centers with both a physician and a nurse pediatric emergency care coordinator had better-than-expected survival, but this association weakened after accounting for trauma center level. Child maltreatment policies were associated with lower-than-expected survival, particularly in Levels III to V trauma centers.

CONCLUSION

Specific components of ED pediatric readiness were associated with pediatric survival among US trauma centers.

LEVEL OF EVIDENCE

Therapeutic/Care Management; Level III.

Emergency Department Pediatric Readiness and Short-term and Long-term Mortality Among Children Receiving Emergency Care

Importance

Emergency departments (EDs) with high pediatric readiness (coordination, personnel, quality improvement, safety, policies, and equipment) are associated with lower mortality among children with critical illness and those admitted to trauma centers, but the benefit among children with more diverse clinical conditions is unknown.

Objective

To evaluate the association between ED pediatric readiness, in-hospital mortality, and 1-year mortality among injured and medically ill children receiving emergency care in 11 states.

Design, Setting, and Participants

This is a retrospective cohort study of children receiving emergency care at 983 EDs in 11 states from January 1, 2012, through December 31, 2017, with follow-up for a subset of children through December 31, 2018. Participants included children younger than 18 years admitted, transferred to another hospital, or dying in the ED, stratified by injury vs medical conditions. Data analysis was performed from November 1, 2021, through June 30, 2022.

Exposure

ED pediatric readiness of the initial ED, measured through the weighted Pediatric Readiness Score (wPRS; range, 0-100) from the 2013 National Pediatric Readiness Project assessment.

Main Outcomes and Measures

The primary outcome was in-hospital mortality, with a secondary outcome of time to death to 1 year among children in 6 states.

Results

There were 796 937 children, including 90 963 (11.4%) in the injury cohort (mean [SD] age, 9.3 [5.8] years; median [IQR] age, 10 [4-15] years; 33 516 [36.8%] female; 1820 [2.0%] deaths) and 705 974 (88.6%) in the medical cohort (mean [SD] age, 5.8 [6.1] years; median [IQR] age, 3 [0-12] years; 329 829 [46.7%] female, 7688 [1.1%] deaths). Among the 983 EDs, the median (IQR) wPRS was 73 (59-87). Compared with EDs in the lowest quartile of ED readiness (quartile 1, wPRS of 0-58), initial care in a quartile 4 ED (wPRS of 88-100) was associated with 60% lower in-hospital mortality among injured children (adjusted odds ratio, 0.40; 95% CI, 0.26-0.60) and 76% lower mortality among medical children (adjusted odds ratio, 0.24; 95% CI, 0.17-0.34). Among 545 921 children followed to 1 year, the adjusted hazard ratio of death in quartile 4 EDs was 0.59 (95% CI, 0.42-0.84) for injured children and 0.34 (95% CI, 0.25-0.45) for medical children. If all EDs were in the highest quartile of pediatric readiness, an estimated 288 injury deaths (95% CI, 281-297 injury deaths) and 1154 medical deaths (95% CI, 1150-1159 medical deaths) may have been prevented.

Conclusions and Relevance

These findings suggest that children with injuries and medical conditions treated in EDs with high pediatric readiness had lower mortality during hospitalization and to 1 year.

National evaluation of the association between stay-at-home orders on mechanism of injury and trauma admission volume

BACKGROUND

The COVID-19 pandemic had numerous negative effects on the US healthcare system. Many states implemented stay-at-home (SAH) orders to slow COVID-19 virus transmission. We measured the association between SAH orders on the injury mechanism type and volume of trauma center admissions during the first wave of the COVID-19 pandemic.

METHODS

All trauma patients aged 16 years and older who were treated at the American College of Surgeons Trauma Quality Improvement Program participating centers from January 2018-September 2020. Weekly trauma patient volume, patient demographics, and injury characteristics were compared across the corresponding SAH time periods from each year. Patient volume was modeled using harmonic regression with a random hospital effect.

RESULTS

There were 166,773 patients admitted in 2020 after a SAH order and an average of 160,962 patients were treated over the corresponding periods in 2018-2019 in 474 centers. Patients presenting with a pre-existing condition of alcohol misuse increased (13,611 (8.3%) vs. 10,440 (6.6%), p <0.001). Assault injuries increased (19,056 (11.4%) vs. 15,605 (9.8%)) and firearm-related injuries (14,246 (8.5%) vs. 10,316 (6.4%)), p<0.001. Firearm-specific assault injuries increased (10,748 (75.5%) vs. 7,600 (74.0%)) as did firearm-specific unintentional injuries (1,318 (9.3%) vs. 830 (8.1%), p<0.001. In the month preceding the SAH orders, trauma center admissions decreased. Within a week of SAH implementation, hospital admissions increased (p<0.001) until a plateau occurred 10 weeks later above predicted levels. On regional sub-analysis, admission volume remained significantly elevated for the Midwest during weeks 11-25 after SAH order implementation, (p<0.001).

ICD-10 based machine learning models outperform the Trauma and Injury Severity Score (TRISS) in survival prediction

Background

Precise models are necessary to estimate mortality risk following traumatic injury to inform clinical decision making or quantify hospital performance. The Trauma and Injury Severity Score (TRISS) has been the historical gold standard in survival prediction but its limitations are well-characterized. The present study used International Classification of Diseases 10^thRevision (ICD-10) injury codes with machine learning approaches to develop models whose performance was compared to that of TRISS.

Methods

The 2015–2017 National Trauma Data Bank was used to identify patients following trauma-related admission. Injury codes from ICD-10 were grouped by clinical relevance into 1,495 variables. The TRISS score, which comprises the Injury Severity Score, age, mechanism (blunt vs penetrating) as well as highest 24-hour values for systolic blood pressure (SBP), respiratory rate (RR) and Glasgow Coma Scale (GCS) was calculated for each patient. A base eXtreme gradient boosting model (XGBoost), a machine learning technique, was developed using injury variables as well as age, SBP, RR, mechanism and GCS. Prediction of in-hospital survival and other in-hospital complications were compared between both models using receiver operating characteristic (ROC) and reliability plots. A complete XGBoost model, containing injury variables, vitals, demographic information and comorbidities, was additionally developed.

Results

Of 1,380,740 patients, 1,338,417 (96.9%) survived to discharge. Compared to survivors, those who died were older and had a greater prevalence of penetrating injuries (18.0% vs 9.44%). The base XGBoost model demonstrated a greater receiver-operating characteristic (ROC) than TRISS (0.950 vs 0.907) which persisted across sub-populations and secondary endpoints. Furthermore, it exhibited high calibration across all risk levels (R² = 0.998 vs 0.816). The complete XGBoost model had an exceptional ROC of 0.960.

Conclusions

We report improved performance of machine learning models over TRISS. Our model may improve stratification of injury severity in clinical and quality improvement settings.

Facts and Fallacy of Benchmark Performance Indicators

Efforts to improve quality in healthcare have arisen from the recognition that the quality of care delivered and resulting outcomes are highly variable. Performance benchmarking using high-quality data to compare risk-adjusted outcomes between hospitals and surgeons has been widely adopted as one means for addressing this problem. In this article we discuss the history, current state, methodologies, and potential pitfalls of benchmarking efforts to improve quality of healthcare in the United States.

Change in outcomes for trauma patients over time: Two decades of a state trauma system

BACKGROUND

Trauma center mortality rates are benchmarked to expected rates of death based on patient and injury characteristics. The expected mortality rate is recalculated from pooled outcomes across a trauma system each year, obscuring system-level change across years. We hypothesized that risk-adjusted mortality would decrease over time within a state-wide trauma system.

METHODS

We identified adult trauma patients presenting to Level I and II Pennsylvania trauma centers, 1999-2018, using the Pennsylvania Trauma Outcomes Study. Multivariable logistic regression generated risk-adjusted models for mortality in all patients, and in key subgroups: penetrating torso injury, blunt multisystem trauma, and patients presenting in shock.

RESULTS

Of 162,646 included patients, 123,518 (76.1%) were white and 108,936 (67.0%) were male. The median age was 49 (interquartile range [IQR] 29-70), median injury severity score was 16 (IQR 10-24), and 87.5% of injuries were blunt. Overall, 9.9% of patients died, and compared to 1999, no year had significantly higher adjusted odds of mortality. Overall mortality was significantly lower in 2007-2009 and 2011-2018. Of patients with blunt, multisystem injuries, 17.7% died, and adjusted mortality improved over time. Mortality rates were 24.9% for penetrating torso injury, and 56.9% for shock, with no significant change. Mortality improved for patients with ISS < 25, but not for the most severely injured.

CONCLUSIONS

Over 20 years, Pennsylvania trauma centers demonstrated improved risk-adjusted mortality rates overall, but improvement remains lacking in high-risk groups despite numerous innovations and practice changes in this time period. Identifying change over time can help guide focus to these critical gaps.

Application of power analysis to determine the optimal reporting time frame for use in statewide trauma system quality reporting

BACKGROUND

Meaningful reporting of quality metrics relies on detecting a statistical difference when a true difference in performance exists. Larger cohorts and longer time frames can produce higher rates of statistical differences. However, older data are less relevant when attempting to enact change in the clinical setting. The selection of time frames must reflect a balance between being too small (type II errors) and too long (stale data). We explored the use of power analysis to optimize time frame selection for trauma quality reporting.

METHODS

Using data from 22 Level III trauma centers, we tested for differences in 4 outcomes within 4 cohorts of patients. With bootstrapping, we calculated the power for rejecting the null hypothesis that no difference exists amongst the centers for different time frames. From the entire sample for each site, we simulated randomly generated datasets. Each simulated dataset was tested for whether a difference was observed from the average. Power was calculated as the percentage of simulated datasets where a difference was observed. This process was repeated for each outcome.

RESULTS

The power calculations for the 4 cohorts revealed that the optimal time frame for Level III trauma centers to assess whether a single site's outcomes are different from the overall average was 2 years based on an 80% cutoff.

CONCLUSION

Power analysis with simulated datasets allows testing of different time frames to assess outcome differences. This type of analysis allows selection of an optimal time frame for benchmarking of Level III trauma center data.

Critically Appraising the Quality of Reporting of American College of Surgeons TQIP Studies in the Era of Large Data Research

BACKGROUND

The American College of Surgeons-Trauma Quality Improvement Program (ACS-TQIP) database is one of the most widely used databases for trauma research. We aimed to critically appraise the quality of the methodological reporting of ACS-TQIP studies.

STUDY DESIGN

The ACS-TQIP bibliography was queried for all studies published between January 2018 and January 2021. The quality of data reporting was assessed using the Strengthening the Reporting of Observational studies in Epidemiology-Reporting of Studies Conducted Using Observational Routinely Collected Health Data (STROBE-RECORD) statement and the JAMA Surgery checklist. Three items from each tool were not applicable and thus excluded. The quality of reporting was compared between high- and low-impact factor (IF) journals (cutoff for high IF is >90th percentile of all surgical journals).

RESULTS

A total of 118 eligible studies were included; 12 (10%) were published in high-IF journals. The median (interquartile range) number of criteria fulfilled was 5 (4-6) for the STROBE-RECORD statement (of 10 items) and 5 (5-6) for the JAMA Surgery checklist (of 7 items). Specifically, 73% of studies did not describe the patient population selection process, 61% did not address data cleaning or the implications of missing values, and 76% did not properly state inclusion/exclusion criteria and/or outcome variables. Studies published in high-IF journals had remarkably higher quality of reporting than those in low-IF journals.

CONCLUSION

The methodological reporting quality of ACS-TQIP studies remains suboptimal. Future efforts should focus on improving adherence to standard reporting guidelines to mitigate potential bias and improve the reproducibility of published studies.

Association of Emergency Department Pediatric Readiness With Mortality to 1 Year Among Injured Children Treated at Trauma Centers

IMPORTANCE

There is substantial variability among emergency departments (EDs) in their readiness to care for acutely ill and injured children, including US trauma centers. While high ED pediatric readiness is associated with improved in-hospital survival among children treated at trauma centers, the association between high ED readiness and long-term outcomes is unknown.

OBJECTIVE

To evaluate the association between ED pediatric readiness and 1-year survival among injured children presenting to 146 trauma centers.

DESIGN, SETTING, AND PARTICIPANTS

In this retrospective cohort study, injured children younger than 18 years who were residents of 8 states with admission, transfer to, or injury-related death at one of 146 participating trauma centers were included. Children cared for in and outside their state of residence were included. Subgroups included those with an Injury Severity Score (ISS) of 16 or more; any Abbreviated Injury Scale (AIS) score of 3 or more; head AIS score of 3 or more; and need for early critical resources. Data were collected from January 2012 to December 2017, with follow-up to December 2018. Data were analyzed from January to July 2021.

EXPOSURES

ED pediatric readiness for the initial ED, measured using the weighted Pediatric Readiness Score (wPRS; range, 0-100) from the 2013 National Pediatric Readiness Project assessment.

MAIN OUTCOMES AND MEASURES

Time to death within 365 days.

RESULTS

Of 88 071 included children, 30 654 (34.8%) were female; 2114 (2.4%) were Asian, 16 730 (10.0%) were Black, and 49 496 (56.2%) were White; and the median (IQR) age was 11 (5-15) years. A total of 1974 (2.2%) died within 1 year of the initial ED visit, including 1768 (2.0%) during hospitalization and 206 (0.2%) following discharge. Subgroups included 12 752 (14.5%) with an ISS of 16 or more, 28 402 (32.2%) with any AIS score of 3 or more, 13 348 (15.2%) with a head AIS of 3 or more, and 9048 (10.3%) requiring early critical resources. Compared with EDs in the lowest wPRS quartile (32-69), children cared for in the highest wPRS quartile (95-100) had lower hazard of death to 1 year (adjusted hazard ratio [aHR], 0.70; 95% CI, 0.56-0.88). Supplemental analyses removing early deaths had similar results (aHR, 0.75; 95% CI, 0.56-0.996). Findings were consistent across subgroups and multiple sensitivity analyses.

CONCLUSIONS AND RELEVANCE

Children treated in high-readiness trauma center EDs after injury had a lower risk of death that persisted to 1 year. High ED readiness is independently associated with long-term survival among injured children.

Alcohol is not associated with increased mortality in adolescent traumatic brain injury patients

PURPOSE

Compared to adults, there is a paucity of data regarding the association of a positive alcohol screen (PAS) and outcomes in adolescent patients with traumatic brain injury (TBI). We hypothesize adolescent TBI patients with a PAS on admission to have increased mortality compared to patients with a negative alcohol screen.

METHODS

The 2017 Trauma Quality Improvement Program database was queried for patients aged 13-17 years presenting with a TBI and serum alcohol screen. Patients with missing information regarding midline shift on imaging and Glasgow Coma Scale (GCS) score were excluded. A multivariable logistic regression analysis for mortality was performed.

RESULTS

From 2553 adolescent TBI patients with an alcohol screen, 220 (8.6%) had a PAS. Median injury severity scores and rates of penetrating trauma (all p > 0.05) were similar between alcohol positive and negative patients. Patients with a PAS had a similar mortality rate (13.2% vs. 12.1%, p = 0.64) compared to patients with a negative screen. Multivariate logistic regression controlling for risk factors associated with mortality revealed a PAS to confer a similar risk of mortality compared to alcohol negative patients (p = 0.40).

CONCLUSION

Adolescent TBI patients with a PAS had similar associated risk of mortality compared to patients with a negative alcohol screen.

All body region injuries are not equal: Differences in pediatric discharge functional status based on Abbreviated Injury Scale (AIS) body regions and severity scores

PURPOSE

Functional outcomes have been proposed for assessing quality of pediatric trauma care. Outcomes assessments often rely on Abbreviated Injury Scale (AIS) severity scores to adjust for injury characteristics, but the relationship between AIS severity and functional impairment is unknown. This study's primary aim was to quantify functional impairment associated with increasing AIS severity scores within body regions. The secondary aim was to assess differences in impairment between body regions based on AIS severity.

METHODS

Children with serious (AIS≥ 3) isolated body region injuries enrolled in a multicenter prospective study were analyzed. The primary outcome was functional status at discharge measured using the Functional Status Scale (FSS). Discharge FSS was compared (1) within each body region across increasing AIS severity scores, and (2) between body regions for injuries with matching AIS scores.

RESULTS

The study included 266 children, with 16% having abnormal FSS at discharge. Worse FSS was associated with increasing AIS severity only for spine injuries. Abnormal FSS was observed in a greater proportion of head injury patients with a severely impaired initial Glasgow Coma Scale (GCS) (GCS< 9) compared to those with a higher GCS score (43% versus 9%; p < 0.01). Patients with AIS 3 extremity and severe head injuries had a higher proportion of abnormal FSS at discharge than AIS 3 abdomen or non-severe head injuries.

CONCLUSIONS

AIS severity does not account for variability in discharge functional impairment within or between body regions. Benchmarking based on functional status assessment requires clinical factors in addition to AIS severity for appropriate risk adjustment.

LEVEL OF EVIDENCE

1 (Prognostic and Epidemiological).

Optimal Out-of-Hospital Blood Pressure in Major Traumatic Brain Injury: A Challenge to the Current Understanding of Hypotension

STUDY OBJECTIVE

Little is known about the out-of-hospital blood pressure ranges associated with optimal outcomes in traumatic brain injuries (TBI). Our objective was to evaluate the associations between out-of-hospital systolic blood pressure (SBP) and multiple hospital outcomes without assuming any predefined thresholds for hypotension, normotension, or hypertension.

METHODS

This was a preplanned secondary analysis from the Excellence in Prehospital Injury Care (EPIC) TBI study. Among patients (age ≥10 years) with major TBIs (Barell Matrix type 1 and/or Abbreviated Injury Scale-head severity ≥3) and lowest out-of-hospital SBPs of 40 to 299 mmHg, we utilized generalized additive models to summarize the distributions of various outcomes as smoothed functions of SBP, adjusting for important and significant confounders. The subjects who were enrolled in the study phase after the out-of-hospital TBI guideline implementation were used to validate the models developed from the preimplementation cohort.

RESULTS

Among 12,169 included cases, the mortality model revealed 3 distinct ranges: (1) a monotonically decreasing relationship between SBP and the adjusted probability of death from 40 to 130 mmHg, (2) lowest adjusted mortality from 130 to 180 mmHg, and (3) rapidly increasing mortality above 180 mmHg. A subanalysis of the cohorts with isolated TBIs and multisystem injuries with TBIs revealed SBP mortality patterns that were similar to each other and to that of the main analysis. While the specific SBP ranges varied somewhat for the nonmortality outcomes (hospital length of stay, ICU length of stay, discharge to skilled nursing/inpatient rehabilitation, and hospital charges), the patterns were very similar to that of mortality. In each model, validation was confirmed utilizing the postimplementation cohort.

CONCLUSION

Optimal adjusted mortality was associated with a surprisingly high SBP range (130 to 180 mmHg). Below this level, there was no point or range of inflection that would indicate a physiologically meaningful threshold for defining hypotension. Nonmortality outcomes showed very similar patterns. These findings highlight how sensitive the injured brain is to compromised perfusion at SBP levels that, heretofore, have been considered adequate or even normal. While the study design does did not allow us to conclude that the currently recommended treatment threshold (<90 mmHg) should be increased, the findings imply that the definition of hypotension in the setting of TBI is too low. Randomized trials evaluating treatment levels significantly higher than 90 mmHg are needed.

Social determinants of health and patient-level mortality prediction after trauma

BACKGROUND

Social determinants of health (SDOH) impact patient outcomes in trauma. Census data are often used to account for SDOH; however, there is no consensus on which variables are most important. Social vulnerability indices offer the advantage of combining multiple constructs into a single variable. Our objective was to determine if incorporation of SDOH in patient-level risk-adjusted outcome modeling improved predictive performance.

METHODS

We evaluated two social vulnerability indices at the zip code level: Distressed Community Index (DCI) and National Risk Index (NRI). Individual variable combinations from Agency for Healthcare Research and Quality's SDOH data set were used for comparison. Patients were obtained from the Pennsylvania Trauma Outcomes Study 2000 to 2020. These measures were added to a validated base mortality prediction model with comparison of area under the curve and Bayesian information criterion. We performed center benchmarking using risk-standardized mortality ratios to evaluate change in rank and outlier status based on SDOH. Geospatial analysis identified geographic variation and autocorrelation.

RESULTS

There were 449,541 patients included. The DCI and NRI were associated with an increase in mortality (adjusted odds ratio, 1.02; 95% confidence interval, 1.01-1.03 per 10% percentile rank increase; p < 0.01, respectively). The DCI, NRI, and seven Agency for Healthcare Research and Quality variables also improved base model fit but discrimination was similar. Two thirds of centers changed mortality ranking when accounting for SDOH compared with the base model alone. Outlier status changed in 7% of centers, most representing an improvement from worse-than-expected to nonoutlier or nonoutlier to better-than-expected. There was significant geographic variation and autocorrelation of the DCI and NRI (DCI; Moran's I 0.62, p = 0.01; NRI; Moran's I 0.34, p = 0.01).

CONCLUSION

Social determinants of health are associated with an individual patient's risk of mortality after injury. Accounting for SDOH may be important in risk adjustment for trauma center benchmarking.

LEVEL OF EVIDENCE

Prognostic/Epidemiologic, level IV.

Extending Trauma Quality Improvement Beyond Trauma Centers: Hospital Variation in Outcomes Among Nontrauma Hospitals

OBJECTIVE

The American College of Surgeons (ACS) conducts a robust quality improvement program for ACS-verified trauma centers, yet many injured patients receive care at non-accredited facilities. This study tested for variation in outcomes across non-trauma hospitals and characterized hospitals associated with increased mortality.

SUMMARY BACKGROUND DATA

The study included state trauma registry data of 37,670 patients treated between January 1, 2013, and December 31, 2015. Clinical data were supplemented with data from the American Hospital Association and US Department of Agriculture, allowing comparisons among 100 nontrauma hospitals.

METHODS

Using Bayesian techniques, risk-adjusted and reliability-adjusted rates of mortality and interfacility transfer, as well as Emergency Departments length-of-stay (ED-LOS) among patients transferred from EDs were calculated for each hospital. Subgroup analyses were performed for patients ages >55 years and those with decreased Glasgow coma scores (GCS). Multiple imputation was used to address missing data.

RESULTS

Mortality varied 3-fold (0.9%-3.1%); interfacility transfer rates varied 46-fold (2.1%-95.6%); and mean ED-LOS varied 3-fold (81-231 minutes). Hospitals that were high and low statistical outliers were identified for each outcome, and subgroup analyses demonstrated comparable hospital variation. Metropolitan hospitals were associated increased mortality [odds ratio (OR) 1.7, P = 0.004], decreased likelihood of interfacility transfer (OR 0.7, P ≤ 0.001), and increased ED-LOS (coef. 0.1, P ≤ 0.001) when compared with nonmetropolitan hospitals and risk-adjusted.

CONCLUSIONS

Wide variation in trauma outcomes exists across nontrauma hospitals. Efforts to improve trauma quality should include engagement of nontrauma hospitals to reduce variation in outcomes of injured patients treated at those facilities.

Concept review of regionalized systems of acute care: Is regionalization the next frontier in sepsis care?

Regionalization has become a buzzword in US health care policy. Regionalization, however, has varied meanings, and definitions have lacked contextual information important to understanding its role in improving care. This concept review is a comprehensive primer and summation of 8 common core components of the national models of regionalization informed by text-based analysis of the writing of involved organizations (professional, regulatory, and research) guided by semistructured interviews with organizational leaders. Further, this generalized model of regionalized care is applied to sepsis care, a novel discussion, drawing on existing small-scale applications. This discussion highlights the fit of regionalization principles to the sepsis care model and the actualized and perceived potential benefits. The principal aim of this concept review is to outline regionalization in the United States and provide a roadmap and novel discussion of regionalized care integration for sepsis care.

OUP accepted manuscript

Background

Risk-adjusted mortality (RAM) analysis and comparisons of clinically relevant subsets of trauma patients allow hospitals to assess performance in different processes of care. The aim of the study was to develop a RAM model and compare RAM ratio (RAMR) in subsets of severely injured adult patients treated in university hospitals (UHs) and emergency hospitals (EHs) in Sweden.

Methods

This was a retrospective study of the Swedish trauma registry data (2013 to 2017) comparing RAMR in patients (aged 15 years or older and New Injury Severity Score (NISS) of more than 15) in the total population (TP) and in multisystem blunt (MB), truncal penetrating (PEN), and severe traumatic brain injury (STBI) subsets treated in UHs and EHs. The RAM model included the variables age, NISS, ASA Physical Status Classification System Score, and physiology on arrival.

Results

In total, 6690 patients were included in the study (4485 from UHs and 2205 from EHs). The logistic regression model showed a good fit. RAMR was 4.0, 3.8, 7.4, and 8.5 percentage points lower in UH versus EH for TP (P < 0.001), MB (P < 0.001), PEN (P = 0.096), and STBI (P = 0.005), respectively. The TP and MB subsets were subgrouped in with (+) and without (−) traumatic brain injury (TBI). RAMR was 7.5 and 7.0, respectively, percentage points lower in UHs than in EHs in TP + TBI and MB + TBI (both P < 0.001). In the TP–TBI (P = 0.027) and MB–TBI (P = 0.107) subsets the RAMR was 1.6 and 1.8 percentage points lower, respectively.

Conclusion

The lower RAMR in UHs versus EH were due to differences in TBI-related mortality. No evidence supported that Swedish EHs provide inferior quality of care for trauma patients without TBI or for patients with penetrating injuries.

Developing a measure of overall intensity of injury care: A latent class analysis

BACKGROUND

While injury is a leading cause of death and debility in older adults, the relationship between intensity of care and trauma remains unknown. The focus of this analysis is to measure the overall intensity of care delivered to injured older adults during hospitalization.

METHODS

We used Centers for Medicare and Medicaid Services Medicare fee-for-service claims data (2013-2014), to identify emergency department-based claims for moderate and severe blunt trauma in age-eligible beneficiaries. Medical procedures associated with care intensity were identified using a modified Delphi method. A latent class model was estimated using the identified procedures, intensive care unit length of stay, demographics, and injury characteristics. Clinical phenotypes for each class were explored.

RESULTS

A total of 683,398 cases were classified as low- (73%), moderate- (23%), and high-intensity care (4%). Greater age and reduced injury severity were indicators of lower intensity, while males, non-Whites, and nonfall mechanisms were more common with high intensity. Intubation/mechanical ventilation was an indicator of high intensity and often occurred with at least one other procedure or an extended intensive care unit stay.

CONCLUSION

This work demonstrates that, although heterogeneous, care for blunt trauma can be evaluated using a single novel measure.

LEVEL OF EVIDENCE

For prognostic/epidemiological studies, level III.

Comparison of a trauma comorbidity index with other measures of comorbidities to estimate risk of trauma mortality

BACKGROUND

Comorbidities influence the outcomes of injured patients, yet a lack of consensus exists regarding how to quantify that association. This study details the development and internal validation of a trauma comorbidity index (TCI) designed for use with trauma registry data and compares its performance to other existing measures to estimate the association between comorbidities and mortality.

METHODS

Indiana state trauma registry data (2013-2015) were used to compare the TCI with the Charlson and Elixhauser comorbidity indices, a count of comorbidities, and comorbidities as separate variables. The TCI approach utilized a randomly selected training cohort and was internally validated in a distinct testing cohort. The C-statistic of the adjusted models was tested using each comorbidity measure in the testing cohort to assess model discrimination. C-statistics were compared using a Wald test, and stratified analyses were performed based on predicted risk of mortality. Multiple imputation was used to address missing data.

RESULTS

The study included 84,903 patients (50% each in training and testing cohorts). The Indiana TCI model demonstrated no significant difference between testing and training cohorts (p = 0.33). It produced a C-statistic of 0.924 in the testing cohort, which was significantly greater than that of models using the other indices (p < 0.05). The C-statistics of models using the Indiana TCI and the inclusion of comorbidities as separate variables-the method used by the American College of Surgeons Trauma Quality Improvement Program-were comparable (p = 0.11) but use of the TCI approach reduced the number of comorbidity-related variables in the mortality model from 19 to one.

CONCLUSIONS

When examining trauma mortality, the TCI approach using Indiana state trauma registry data demonstrated superior model discrimination and/or parsimony compared to other measures of comorbidities.

Machine learning for outcome predictions of patients with trauma during emergency department care

OBJECTIVES

To develop and evaluate a machine learning model for predicting patient with trauma mortality within the US emergency departments.

METHODS

This was a retrospective prognostic study using deidentified patient visit data from years 2007 to 2014 of the National Trauma Data Bank. The predictive model intelligence building process is designed based on patient demographics, vital signs, comorbid conditions, arrival mode and hospital transfer status. The mortality prediction model was evaluated on its sensitivity, specificity, area under receiver operating curve (AUC), positive and negative predictive value, and Matthews correlation coefficient.

RESULTS

Our final dataset consisted of 2 007 485 patient visits (36.45% female, mean age of 45), 8198 (0.4%) of which resulted in mortality. Our model achieved AUC and sensitivity-specificity gap of 0.86 (95% CI 0.85 to 0.87), 0.44 for children and 0.85 (95% CI 0.85 to 0.85), 0.44 for adults. The all ages model characteristics indicate it generalised, with an AUC and gap of 0.85 (95% CI 0.85 to 0.85), 0.45. Excluding fall injuries weakened the child model (AUC 0.85, 95% CI 0.84 to 0.86) but strengthened adult (AUC 0.87, 95% CI 0.87 to 0.87) and all ages (AUC 0.86, 95% CI 0.86 to 0.86) models.

CONCLUSIONS

Our machine learning model demonstrates similar performance to contemporary machine learning models without requiring restrictive criteria or extensive medical expertise. These results suggest that machine learning models for trauma outcome prediction can generalise to patients with trauma across the USA and may be able to provide decision support to medical providers in any healthcare setting.