Mehreen Kisat

Influence of the National Trauma Data Bank on the Study of Trauma Outcomes: Is it Time to Set Research Best Practices to Further Enhance Its Impact?

BACKGROUND Risk-adjusted analyses are critical in evaluating trauma outcomes. The National Trauma Data Bank (NTDB) is a statistically robust registry that allows such analyses; however, analytical techniques are not yet standardized. In this study, we examined peer-reviewed manuscripts published using NTDB data, with particular attention to characteristics strongly associated with trauma outcomes. Our objective was to determine if there are substantial variations in the methodology and quality of risk-adjusted analyses and therefore, whether development of best practices for risk-adjusted analyses is warranted. STUDY DESIGN A database of all studies using NTDB data published through December 2010 was created by searching PubMed and Embase. Studies with multivariate risk-adjusted analyses were examined for their central question, main outcomes measures, analytical techniques, covariates in adjusted analyses, and handling of missing data. RESULTS Of 286 NTDB publications, 122 performed a multivariable adjusted analysis. These studies focused on clinical outcomes (51 studies), public health policy or injury prevention (30), quality (16), disparities (15), trauma center designation (6), or scoring systems (4). Mortality was the main outcome in 98 of these studies. There were considerable differences in the covariates used for case adjustment. The 3 covariates most frequently controlled for were age (95%), Injury Severity Score (85%), and sex (78%). Up to 43% of studies did not control for the 5 basic covariates necessary to conduct a risk-adjusted analysis of trauma mortality. Less than 10% of studies used clustering to adjust for facility differences or imputation to handle missing data. CONCLUSIONS There is significant variability in how risk-adjusted analyses using data from the NTDB are performed. Best practices are needed to further improve the quality of research from the NTDB.

Outcome of selective non-operative management of penetrating abdominal injuries from the North American National Trauma Database

The aim of this study was to investigate trends in the practice of selective non‐operative management (SNOM) for penetrating abdominal injury (PAI) and to determine factors associated with its failure.

Hospital-based trauma quality improvement initiatives: first step toward improving trauma outcomes in the developing world.

BACKGROUND Injuries remain a leading cause of death in the developing world. Whereas new investments are welcome, quality improvement (QI) at the currently available trauma care facilities is essential. The objective of this study was to determine the effect and long-term sustainability of trauma QI initiatives on in-hospital mortality and complications at a large tertiary hospital in a developing country. METHODS In 2002, a specialized trauma team was formed (members trained using advanced trauma life support), and a western style trauma program established including a registry and quality assurance program. Patients from 1998 onward were entered in to this registry, enabling a preimplementation and postimplementation study. Adults (>15 years) with blunt or penetrating trauma were analyzed. The main outcomes of interest were (1) in-hospital mortality and (2) occurrence of any complication. Multiple logistic regression was performed to assess the impact of formalized trauma care on outcomes, controlling for covariates reaching significance in the bivariate analyses. RESULTS A total of 1,227 patient records were analyzed. Patient demographics and injury characteristics are described in Table 1. Overall in-hospital mortality rate was 6.4%, and the complication rate was 11.1%. On multivariate analysis, patients admitted during the trauma service years were 4.9 times less likely to die (95% confidence interval, 1.77–13.57) and 2.60 times (odds ratio; 95% confidence interval, 1.29–5.21) less likely to have a complication compared with those treated in the pretrauma service years. CONCLUSION Despite significant delays in hospital transit and lack of prehospital trauma care, hospital level implementation of trauma QI program greatly decreases mortality and complication rates in the developing world. LEVEL OF EVIDENCE Care management study, level IV.

Predictors of sepsis in moderately severely injured patients: an analysis of the National Trauma Data Bank

BACKGROUND Post-traumatic sepsis is a significant cause of in-hospital death. However, socio-demographic and clinical characteristics that may predict sepsis in injured patients are not well known. The objective of this study was to identify risk factors that may be associated with post-traumatic sepsis. METHODS Retrospective analysis of patients in the National Trauma Data Bank for 2007-2008. Patients older than 16 years of age with an Injury Severity Score (ISS) ≥ 9 points were included. Multivariable logistic regression was used to determine association of sepsis with patient (age, gender, ethnicity, and insurance status), injury (mechanism, ISS, injury type, hypotension), and clinical (major surgical procedure, intensive care unit admission) characteristics. RESULTS Of a total of 1.3 million patients, 373,370 met the study criteria, and 1.4% developed sepsis, with an associated mortality rate of approximately 20%. Age, male gender, African-American race, hypotension on emergency department presentation, and motor vehicle crash as the injury mechanism were independently associated with post-traumatic sepsis. CONCLUSIONS Socio-demographic and injury factors, such as age, race, hypotension on admission, and severity and mechanism of injury predict post-traumatic sepsis significantly. Further exploration to explain why these patient groups are at increased risk is warranted in order to understand better and potentially prevent this life-threatening complication.

Increased mortality associated with EMS transport of gunshot wound victims when compared to private vehicle transport

BACKGROUND Recent studies suggest that mode of transport affects survival in penetrating trauma patients. We hypothesised that there is wide variation in transport mode for patients with gunshot wounds (GSW) and there may be a mortality difference for GSW patients transported by emergency medical services (EMS) vs. private vehicle (PV). STUDY DESIGN We studied adult (≥16 years) GSW patients in the National Trauma Data Bank (2007-2010). Level 1 and 2 trauma centres (TC) receiving ≥50 GSW patients per year were included. Proportions of patients arriving by each transport mode for each TC were examined. In-hospital mortality was compared between the two groups, PV and EMS, using multivariable regression analyses. Models were adjusted for patient demographics, injury severity, and were adjusted for clustering by facility. RESULTS 74,187 GSW patients were treated at 182 TCs. The majority (76%) were transported by EMS while 12.6% were transported by PV. By individual TC, the proportion of patients transported by each category varied widely: EMS (median 78%, interquartile range (IQR) 66-85%), PV (median 11%, IQR 7-17%), or others (median 7%, IQR 2-18%). Unadjusted mortality was significantly different between PV and EMS (2.1% vs. 9.7%, p<0.001). Multivariable analysis demonstrated that EMS transported patients had a greater than twofold odds of dying when compared to PV (OR=2.0, 95% CI 1.73-2.35). CONCLUSIONS Wide variation exists in transport mode for GSW patients across the United States. Mortality may be higher for GSW patients transported by EMS when compared to private vehicle transport. Further studies should be performed to examine this question.

Disparities in mortality after blunt injury: Does insurance type matter?

BACKGROUND Insurance-related outcomes disparities are well-known, but associations between distinct insurance types and trauma outcomes remain unclear. Prior studies have generally merged various insurance types into broad groups. The purpose of this study is to determine the association of specific insurance types with mortality after blunt injury. MATERIALS AND METHODS Cases of blunt injury among adults aged 18-64 y with an injury severity score >9 were identified using the 2007-2009 National Trauma Data Bank. Crude mortality was calculated for 10 insurance types. Multivariable logistic regression was employed to determine difference in odds of death between insurance types, controlling for injury severity score, Glasgow Coma Scale motor, mechanism of injury, sex, race, and hypotension. Clustering was used to account for possible inter-facility variations. RESULTS A total of 312,312 cases met inclusion criteria. Crude mortality ranged from 3.2 to 6.0% by insurance type. Private Insurance, Blue Cross Blue Shield, Workers Compensation, and Medicaid yielded the lowest relative odds of death, while Not Billed and Self Pay yielded the highest. Compared with Private Insurance, odds of death were higher for No Fault (OR 1.25, P = 0.022), Not Billed (OR 1.77, P < 0.001), and Self Pay (OR 1.77, P < 0.001). Odds of death were higher for Medicare (OR 1.52, P < 0.001) and Other Government (OR 1.35, P = 0.049), while odds of death were lower for Medicaid (OR 0.89, P = 0.015). CONCLUSIONS Significant differences in mortality after blunt injury were seen between insurance types, even among those commonly grouped in other studies. Policymakers may use this information to implement programs to monitor and reduce insurance-related disparities.

Association between intentional injury and long-term survival after trauma

Objective:To determine the risk-adjusted mortality of intentionally injured patients within 7 to 9 years postinjury, compared with unintentionally injured patients. Background:Violent injury contributes significantly to trauma mortality in the United States. Homicide is the second leading killer of American youth, aged 15 to 24 years. Long-term survival among intentionally injured patients has not been well studied. It is also unknown whether intentionally injured patients have worse long-term survival compared with unintentionally or accidentally injured patients with equivalent injuries. Methods:Adult trauma patients admitted for 24 hours or more and discharged alive from the Johns Hopkins Hospital from January 1, 1998, to December 31, 2000, were included. The primary outcome was mortality within 7 to 9 years postinjury. Long-term patient survival was determined using the National Death Index. The association between injury intentionality and mortality was investigated using a Cox proportional hazard regression model, adjusted for confounders such as injury severity and patient race, socioeconomic status, and comorbid conditions. Overall differences in survival between those with intentional versus unintentional injury were also determined by comparing adjusted Kaplan-Meier survival curves. Results:A total of 2062 patients met inclusion criteria. Of these, 56.4% were intentionally injured and 43.6% were unintentionally injured. Compared with unintentionally injured patients, intentionally injured patients were younger and more often male and from a zip code with low median household income. Approximately 15% of all patients had died within 7 to 9 years of follow-up. Older age and presence of comorbidities were associated with this outcome; however, intentional injury was not found to be significantly associated with long-term mortality rates. There was also no significant difference in survival curves between groups; intentionally injured patients were much more likely to die of a subsequent injury, whereas those with unintentional injury commonly died of noninjury causes. Conclusions:There was no significant difference in mortality between intentionally injured and unintentionally injured patients within 7 to 9 years postinjury. These results confirm the long-term effectiveness of lifesaving trauma care for those with intentional injury. However, given that patients with intentional injuries were more likely to suffer a subsequent violent death, interventions focused on breaking the cycle of violence are needed.

Predictors of positive head CT scan and neurosurgical procedures after minor head trauma

BACKGROUND There continues to be an ongoing debate regarding the utility of head CT scans in patients with a normal Glasgow Coma Scale (GCS) after minor head injury. The objective of this study is to determine patient and injury characteristics that predict a positive head CT scan or need for a neurosurgical procedure (NSP) among patients with blunt head injury and a normal GCS. MATERIALS AND METHODS Retrospective analysis of adult patients in the National Trauma Data Bank who presented to the ED with a history of blunt head injury and a normal GCS of 15. The primary outcomes were a positive head CT scan or a NSP. Multivariate logistic regression controlling for patient and injury characteristics was used to determine predictors of each outcome. RESULTS Out of a total of 83,566 patients, 24,414 (29.2%) had a positive head CT scan and 3476 (4.2%) underwent a NSP. Older patients and patients with a history of fall (compared with a motor vehicle crash) were more likely to have a positive finding on a head CT scan. Male patients, African-Americans (compared with Caucasians), and those who presented with a fall were more likely to have a NSP. CONCLUSIONS Older age, male gender, ethnicity, and mechanism of injury are significant predictors of a positive finding on head CT scans and the need for neurosurgical procedures. This study highlights patient and injury-specific characteristics that may help in identifying patients with supposedly minor head injury who will benefit from a head CT scan.

Beyond incidence: Costs of complications in trauma and what it means for those who pay

INTRODUCTION Trauma patients have greater rates of complications than general surgery patients; however, existing surgical pay-for-performance (P4P) guidelines have yet to be adapted for trauma care. To better understand whether current P4P measures are applicable to trauma, this study used nationally representative data to determine the mortality and attributable costs associated with the presence or absence of both Centers for Medicare and Medicaid Services-recognized complications (urinary tract infections, surgical site infections [SSIs], and pneumonia) and other major trauma-related complications. METHODS Trauma admissions were extracted from the 2008 National Inpatient Sample using primary ICD-9-CM diagnosis codes (range, 800-905, 910-939, 950-958). Patients aged 18-65 years with a duration of hospital stay of >3 days and isolated complications were included. To account for differences in patient factors, coarsened-exact matching was used to create comparable cohorts of adult patients with and without complications. Multivariable regression was then performed within matched groups to determine differences in cost and mortality, controlling for hospital characteristics and wage index. RESULTS Of 493,372 trauma patients, 78,156 met inclusion criteria, of whom 24.4% had an isolated complication. Consistent with surgical P4P guidelines, SSI, urinary tract infections, and pneumonia had the greatest incidence (8.0%, 5.2%, and 4.4%, respectively); however, mortality in matched patients with complications was greatest for sepsis (odds ratio [OR], 9.76; 95% CI, 3.84-24.80), myocardial infarction (MI; OR, 4.21; 95% CI, 1.70-10.44) and stroke (OR, 3.02; 95% CI, 1.40-6.52). Excess costs associated with a complication were similarly greatest for sepsis (relative cost, 1.84; 95% CI, 1.57-2.17), followed by acute respiratory distress syndrome (ARDS; relative cost, 1.84; 95% CI, 1.7-1.99) and MI (relative cost, 1.73; 95% CI, 1.51-1.99). CONCLUSION Consideration of attributable costs and mortality suggest that additional complications have a substantial impact among trauma patients, beyond the conditions used in general surgery P4P guidelines. These aspects of trauma should be prioritized to capture the influence of complications in trauma that the incidence of frequent but less costly conditions overlooks.

State-by-state variation in emergency versus elective colon resections: room for improvement.

BACKGROUND: Compared with elective surgical procedures, emergency procedures are associated with higher cost, morbidity, and mortality. This study seeks to investigate potential state‐by‐state variations in the incidence of emergent versus elective colon resections. METHODS: A retrospective analysis of all adult patients (aged >=18 years) included in the Nationwide Inpatient Sample from 2005 to 2009 who underwent hemicolectomy (right or left) or sigmoidectomy was conducted. Discharge‐level weights were applied, and generalized linear models were used to assess the odds of a patient undergoing emergent versus elective colon surgery nationally and for each state after adjusting for patient and hospital factors. Odds ratios (ORs) were estimated with the national average as the reference. RESULTS: The final study cohort included 203,050 observations composed of 83,090 emergent and 119,960 elective colectomies. The state with the highest unadjusted proportion of emergent procedures was Nevada (53.6%), whereas Texas had the lowest (2.8%). Compared with the national average, the adjusted odds of undergoing emergency colectomy remained highest in Nevada (OR, 1.70; 95% confidence interval, 1.54–1.87) and lowest in Texas (OR, 0.43; 95% confidence interval, 0.36–0.51). CONCLUSION: Substantial state variations exist in rates of emergency colon surgery within the United States. Identification of these differences suggests significant variations in practice and a potential to decrease the number of emergent colon operations. LEVEL OF EVIDENCE: Prognostic and epidemiologic study, level III.