Social Determinants of Health in Trauma.

Social determinants (SD) refer to a variety of environmental factors that can influence certain clinical outcomes. SD that affect clinical outcomes in trauma patients are largely understudied. We hypothesized that patients with an "at risk" (AR) classification for any SD will have a greater frequency of negative outcomes when compared to their "not at risk" (NAR) counterparts. A retrospective review was performed (1/2021-2/2022) of all trauma patients that met the inclusion criteria. 2225 patients were included. SD included (based on collection rates) for analysis were: tobacco use, stress, and intimate partner violence. Tobacco usage was the only SD that was significantly associated with a higher 30-day mortality. This study demonstrated a paucity of data in the medical record regarding SD. In an effort to provide more inclusive care and address health disparities in our patient population, there is a need for more complete data collection upon admission/duration of hospital stay.

Toward Predicting 30-Day Readmission Among Oncology Patients: Identifying Timely and Actionable Risk Factors.

PURPOSE: Predicting 30-day readmission risk is paramount to improving the quality of patient care. In this study, we compare sets of patient-, provider-, and community-level variables that are available at two different points of a patient's inpatient encounter (first 48 hours and the full encounter) to train readmission prediction models and identify possible targets for appropriate interventions that can potentially reduce avoidable readmissions. METHODS: Using electronic health record data from a retrospective cohort of 2,460 oncology patients and a comprehensive machine learning analysis pipeline, we trained and tested models predicting 30-day readmission on the basis of data available within the first 48 hours of admission and from the entire hospital encounter. RESULTS: Leveraging all features, the light gradient boosting model produced higher, but comparable performance (area under receiver operating characteristic curve [AUROC]: 0.711) with the Epic model (AUROC: 0.697). Given features in the first 48 hours, the random forest model produces higher AUROC (0.684) than the Epic model (AUROC: 0.676). Both models flagged patients with a similar distribution of race and sex; however, our light gradient boosting and random forest models were more inclusive, flagging more patients among younger age groups. The Epic models were more sensitive to identifying patients with an average lower zip income. Our 48-hour models were powered by novel features at various levels: patient (weight change over 365 days, depression symptoms, laboratory values, and cancer type), hospital (winter discharge and hospital admission type), and community (zip income and marital status of partner). CONCLUSION: We developed and validated models comparable with the existing Epic 30-day readmission models with several novel actionable insights that could create service interventions deployed by the case management or discharge planning teams that may decrease readmission rates over time.

The Homeless Population: Medical Refugees in a Mature Trauma System.

INTRODUCTION: On any given day, there are >550,000 homeless persons in the United States. Little research has examined the relationship between the homeless population and traumatic injuries. We hypothesized that homeless trauma patients have a higher mortality compared to those who are not homeless. METHODS: The Pennsylvania Trauma Outcome Study database was retrospectively queried from 2003-2018 for all adult (age ≥15) patients admitted to trauma centers in Pennsylvania. Homelessness was defined as homeless on admission or homeless as their discharge status. Patient demographics, comorbidities, and clinical variables were compared between homeless and non-homeless patients. Logistic regression was used to control for age, gender, injury severity, injury type, admission Glasgow Coma Scale, and systolic blood pressure to assess morbidity and mortality. RESULTS: 773 patients were identified upon query. Homeless trauma patients were more likely to be male, younger, black, and of Hispanic ethnicity. Compared to non-homeless, they were more likely to have a positive drug screen or mental illness at the time of injury. They were not more significantly injured than their counterparts; however, in adjusted analysis, the homeless had significantly higher odds of both complications (Adjusted Odds Ratio [AOR]: 3.11; 95%CI: 2.64-3.66, P < .001) and mortality (AOR: 1.79; 95%CI: 1.29-2.50, P = .001). CONCLUSION: Although homeless patients were not more severely injured than the general trauma population, they had significantly higher odds of both complications and mortality. This population represents a very vulnerable community in need of medical intervention and injury prevention programs.

A Pilot of a Randomized Control Trial of Melatonin and Vitamin C for Mild-to-Moderate COVID-19.

This study aimed to help determine the effect of dietary supplements on symptom course and quality of life in patients with mild-to-moderate COVID-19 infection. DESIGN: We modified the Wisconsin Upper Respiratory Symptom Survey (WURSS) to conduct a 3 arm, parallel, randomized, double-blind, placebo-controlled trial, enrolling patients with mild-to-moderate symptoms of COVID-19 infection. Patients took placebo (n = 34), vitamin C 1000 mg (n = 32), or melatonin 10 mg (n = 32) orally for 14 days. OUTCOMES: Ninety Eight (98 out of 104 recruited; mean age = 52 years) patients completed the study. Outcomes were calculated as differences from baseline scores on each of 2 WURSS-derived surveys and analyzed using a spline regression analysis. Regarding symptom progression, those patients taking placebo and vitamin C progressed at the same rate. When compared with those taking placebo (coefficient = -1.09 (95% confidence interval [CI] = -1.39 to -0.8) the group taking melatonin had a faster resolution of symptoms (coefficient = -0.63 [95% CI -1.02 to -0.21] P = .003). By day 14 all 3 groups had reached plateau.Quality-of-life impact analysis demonstrated that the group taking vitamin C improved at the same rate as the group taking placebo (coefficient = -0.71 (95% CI = -1.11 to -0.3)). The group taking melatonin (coefficient = -1.16 (95% CI = -1.75 to - 0.57) P < .005) had a faster improvement in quality-of-life. By day 14 all 3 groups had reached plateau. CONCLUSION: Vitamin C 1000 mg once daily has no effect on disease progression. Melatonin 10 mg daily may have a statistically significant effect but it is unclear if this represents a clinically significant benefit to those with mild-to-moderate symptoms of COVID-19 infection. Further study is warranted.

Location of Injury and Other Factors Associated With Increased Survival Among Severely Injured Geriatric Trauma Patients.

BACKGROUND: Patients with a Trauma Injury Severity Score (TRISS) < .5 are termed "unexpected survivors." There is scarce information published on this subset of geriatric patients whose survival is an anomaly. METHODS: This is a retrospective case-control study examining all geriatric patients (age ≥65) not expected to survive (TRISS<.5) in the Pennsylvania Trauma Outcome Study (PTOS) database from 2013 to 2017. Primary outcome was survival to discharge. We selected 10 clinically important variables for logistic regression analysis as possible factors that may improve survival. RESULTS: 1336 patients were included, 395 (29.6%) were unexpected survivors. Factors that improved survival odds are the following: Place of injury: street/highway (AOR:0.51; 95% CI: .36-.73, P < .001) and residential institution (AOR:0.46; 95% CI: .21-.98, P = .043); and presence of Benzodiazepines (AOR:0.49; 95% CI: .31-.77, P = .002) or ethanol (AOR:0.57; 95% CI: .34-.97, P = .040). Factors that decreased survival odds are the following: Hypotension (AOR: 8.59; 95% CI: 4.33-17.01, P < .001) and hypothermia (AOR: 1.58; 95% CI: 1.10-2.28, P = .014). Gender, race/ethnicity, blood transfusion in first 24 hours, shift of presentation to Emergency Department, place of injury (farm, industrial, recreational, or public building), use of Tetrahydrocannabinol, amphetamines or opioids, and level of trauma activation did not impact survival. DISCUSSION: Location of injury (street/highway and residential institution) and ethanol or benzodiazepine use led to a significant increased survival in severely injured geriatric patients. Hypotension and hypothermia led to decreased survival. Future studies should determine possible reasons these factors lead to survival (and identify additional factors) to focus efforts in these areas to improve outcomes in geriatric trauma.

An Evaluation of Pediatric Secondary Overtriage in the Pennsylvania Trauma System.

BACKGROUND: We sought to determine the secondary overtriage rate of pediatric trauma patients admitted to pediatric trauma centers. We hypothesized that pediatric secondary overtriage (POT) would constitute a large percentage of admissions to PTC. MATERIALS AND METHODS: The Pennsylvania Trauma Outcome Study database was retrospectively queried from 2003 to 2017 for pediatric (age ≤ 18 y) trauma patients transferred to accredited pediatric trauma centers in Pennsylvania (n = 6). Patients were stratified based on discharge within (early) and beyond (late) 24 h following admission. POT was defined as patients transferred to a PTC with an early discharge. Multilevel mixed-effects logistic regression model controlling for demographic and injury severity covariates were utilized to determine the adjusted impact of injury patterns on early discharge. RESULTS: A total of 37,653 patients met inclusion criteria. For transfers, POT compromised 18,752 (49.8%) patients. Compared to POT, non-POT were more severely injured (ISS: 10 versus 6;P < 0.001) and spent less time in the ED (Min: 181 versus 207;P < 0.001). In adjusted analysis, concussion, closed skull vault fractures, supracondylar humerus fractures, and consults to neurosurgery were associated with increased odds of POT. Overall, femur fracture, child abuse evaluation, and consults to plastic surgery, orthopedics, and ophthalmology were all associated with a decreased risk of being POT. CONCLUSIONS: POT comprises 49.8% of PTC transfer admissions in Pennsylvania's trauma system. Improving community resources for management of pediatric concussion and mild TBI could result in decreased rates of POT to PTCs. Developing better inter-facility transfer guidelines and increased education of adult TC and nontrauma center hospitals is needed to decrease POT. LEVEL OF EVIDENCE: Epidemiologic study, level III.

The effect of the Affordable Care Act on rates on inpatient rehabilitation hospital admission in a mature trauma system.

BACKGROUND: The beneficial effects of acute rehabilitation for trauma patients are well documented but can be limited because of insurance coverage. The Patient Protection and Affordable Care Act (ACA) went into effect on March 23, 2010. The ACA allowed patients who previously did not have insurance to be fully incorporated into the health system. We sought to analyze the likelihood of discharge to rehab for trauma patients before and after the implementation of the ACA. We hypothesized that there would be a higher rate of inpatient rehabilitation hospital (IRH) admission after the ACA was put into effect. METHODS: The Pennsylvania Trauma Outcome Study (PTOS) database was retrospectively queried from 2003 to 2017 for all trauma patients admitted to accredited trauma centers in Pennsylvania, who also had a functional status at discharge (FSD). Admission to an IRH was determined using discharge destination. Two categories were created to represent periods before and after ACA was implemented, 2003 to 2009 (pre-ACA) and 2010-2017 (post-ACA). A multilevel mixed-effects logistic regression model controlling for demographics, injury severity, and FSD assessed the adjusted impact of ACA implementation on IRH admissions. RESULTS: From the Pennsylvania Trauma Outcome Study query, 341,252 patients had FSD scores and of these patients, 47,522 (13.9%) were admitted to IRH. Patients who were severely injured were more likely to be admitted to IRH. Compared with FSD scores signifying complete independence at discharge, those with lower FSD had significantly increased odds of IRH admission. The odds of IRH admission post-ACA implementation significantly increased when compared with pre-ACA years (adjusted odds ratio, 1.14; 95% confidence interval, 1.12-1.17; p < 0.001; area under the receiver operating curve, 0.818). CONCLUSION: The implementation of the ACA significantly increased the likelihood of discharge to IRH for trauma patients. LEVEL OF EVIDENCE: Care management, level III.

Geriatric Trauma Mortality: Does Trauma Center Level Matter?

BACKGROUND: Given their mostly rural/suburban locations, level II trauma centers (TCs) may offer greater exposure to and experience in managing geriatric trauma patients. We hypothesized that geriatric patients would have improved outcomes at level II TCs compared to level I TCs. METHODS: The Pennsylvania Trauma Outcome Study (PTOS) database was retrospectively queried from 2003 to 2017 for geriatric (age ≥65 years) trauma patients admitted to level I and II TCs in Pennsylvania. Patient demographics, injury severity, and clinical outcomes were compared to assess differences in care between level I and II TCs. A multivariate logistic regression model assessed the adjusted impact of care at level I vs II TCs on mortality, complications, and functional status at discharge (FSD). The National Trauma Data Bank (NTDB) was retrospectively queried for geriatric (age ≥65 years) trauma admissions to state-accredited level I or level II TCs in 2013. RESULTS: 112 648 patients met inclusion criteria. The proportion of geriatric trauma patients across level I and level II TCs were determined to be 29.1% and 36.2% (P <.001), respectively. In adjusted analysis, there was no difference in mortality (adjusted odds ratio [AOR]: 1.13; P = .375), complications (AOR: 1.25; P = .080) or FSD (AOR: 1.09; P = .493) when comparing level I to level II TCs. Adjusted analysis from the NTDB (n = 144 622) also found that mortality was not associated with TC level (AOR: 1.04; P = .182). DISCUSSION: Level I and level II TCs had similar rates of mortality, complications, and functional outcomes despite a higher proportion (but lower absolute number) of geriatric patients being admitted to level II TCs. Future consideration for location of centers of excellence in geriatric trauma should include both level I and II TCs.

An analysis of overtriage and undertriage by advanced life support transport in a mature trauma system.

BACKGROUND: While issues regarding triage of severely injured trauma patients are well publicized, little information exists concerning the difference between triage rates for patients transported by advanced life support (ALS) and basic life support (BLS). We sought to analyze statewide trends in undertriage (UT) and overtriage (OT) to address this question, hypothesizing that there would be a difference between the UT and OT rates for ALS compared with BLS over a 13-year period. METHODS: All patients submitted to Pennsylvania Trauma Outcomes Study database from 2003 to 2015 were analyzed. Undertriage was defined as not calling a trauma alert for patients with an Injury Severity Score (ISS) of 16 or greater. Overtriage was defined as calling a trauma alert for patients with an ISS of 9 or less. A logistic regression was used to assess mortality between triage groups in ALS and BLS. A multinomial logistic regression assessed the adjusted impact of ALS versus BLS transport on UT and OT versus normal triage while controlling for age, sex, Glasgow Coma Scale, systolic blood pressure (SBP), pulse, Shock Index and injury year. RESULTS: A total of 462,830 patients met inclusion criteria, of which 115,825 had an ISS of 16 or greater and 257,855 had an ISS of 9 or less. Both ALS and BLS had significantly increased mortality when patients were undertriaged compared with the reference group. Multivariate analysis in the form of a multinomial logistic regression revealed that patients transported by ALS had a decreased adjusted rate of undertriage (relative risk ratio, 0.92; 95% confidence interval, 0.87-0.97; p = 0.003) and an increased adjusted rate of OT (relative risk ratio, 1.59; 95% confidence interval, 1.54-1.64; p < 0.001) compared with patients transported by BLS. CONCLUSION: Compared with their BLS counterparts, while UT is significantly lower, OT is substantially higher in ALS-further increasing the high levels of resource (over)utilization in trauma patients. Undertriage in both ALS and BLS are associated with increased mortality rates. Additional education, especially in the BLS provider, on identifying the major trauma victim may be warranted based on the results of this study. LEVEL OF EVIDENCE: Epidemiological, Level III.