Michael A. Horst

Trends in broad-spectrum antibiotic prescribing for children with acute otitis media in the United States, 1998–2004

BackgroundOveruse of broad-spectrum antibiotics is associated with antibiotic resistance. Acute otitis media (AOM) is responsible for a large proportion of antibiotics prescribed for US children. Rates of broad-spectrum antibiotic prescribing for AOM are unknown.MethodsAnalysis of the National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey, 1998 to 2004 (N = 6,878). Setting is office-based physicians, hospital outpatient departments, and emergency departments. Patients are children aged 12 years and younger prescribed antibiotics for acute otitis media. Main outcome measure is percentage of broad-spectrum antibiotics, defined as amoxicillin/clavulanate, macrolides, cephalosporins and quinolones.ResultsBroad-spectrum prescribing for acute otitis media increased from 34% of visits in 1998 to 45% of visits in 2004 (P < .001 for trend). The trend was primarily attributable to an increase in prescribing of amoxicillin/clavulanate (8% to 15%; P < .001 for trend) and macrolides (9% to 15%; P < .001 for trend). Prescribing remained stable for amoxicillin and cephalosporins while decreasing for narrow-spectrum agents (12% to 3%; P < .001 for trend) over the study period. Independent predictors of broad-spectrum antibiotic prescribing were ear pain, non-white race, public and other insurance (compared to private), hospital outpatient department setting, emergency department setting, and West region (compared to South and Midwest regions), each of which was associated with lower rates of broad-spectrum prescribing. Age and fever were not associated with prescribing choice.ConclusionPrescribing of broad-spectrum antibiotics for acute otitis media has steadily increased from 1998 to 2004. Associations with non-clinical factors suggest potential for improvement in prescribing practice.

High-risk geriatric protocol: improving mortality in the elderly.

BACKGROUND Injured geriatric patients pose unique challenges to the trauma team because of their abnormal responses to shock and injury. We have developed the high-risk geriatric protocol (GP) that seeks to identify high-risk geriatric patients. We hypothesized that a high-risk GP would improve outcome in this select group of patients. METHODS Patients from 2000 to 2010 were included. Patients 65 years or older who met high-risk GP based on comorbidities and/or physiologic parameters were compared with those patients who had not received GP before its implementation as well as other non-GP patients. This protocol includes a geriatric consultation, as well as a lactate levels, arterial blood gas levels, and echo test to assess for occult shock. Age, trauma activation, preexisting conditions, Injury Severity Score, Revised Trauma Score, and mortality were reviewed. Univariate and multivariate analyses were conducted to identify factors predictive of mortality. RESULTS A total of 3,902 patients were evaluated. Patients receiving GP were less likely to die (odds ratio, 0.63 [0.39–0.99], p = 0.046). For all patients, there was a dramatic increase in mortality for those patients older than 75 years. CONCLUSION The GP, adjusted for other covariates, significantly reduced mortality in our patient population. Thus, this study confirms the overall effectiveness of our GP, which is hallmarked by prompt identification of those patients with occult shock and a multidisciplinary care of the aged population. LEVEL OF EVIDENCE Therapeutic study, level IV.

Factors associated with patient satisfaction scores for physician care in trauma patients.

BACKGROUND The Affordable Care Act of 2010 identifies “patient experience of care” as one of five domains of excellent care. We hypothesized that there are specific demographic factors associated with higher or lower physician satisfaction (PS) scores in trauma patients. METHODS Press-Ganey PS scores for September 2004 to December 2010 were compared with trauma variables and the association of a mean PS greater than or equal to 75 (high score) or less than or equal to 50 (low score). Those variables that proved significant on univariate analysis were subjected to multivariate logistic regression analysis. Significance was at p < 0.05. RESULTS There were 12,196 admissions, of whom 1,631 (13.4%) returned patient satisfaction survey. A total of 1,174 patients (75.5%) returned a high PS (≥75), and 126 patients (8.1%) returned a low PS (⩽50). In the multiple logistic regression analysis, 65 years or older (odds ratio [OR], 1.7), having had a surgical procedure (OR, 1.6), and having a positive impression of the hospital care (OR, 7.0) proved significant for a high PS. Those patients who scored a low PS were significantly more likely to be younger (18–29 years: OR, 2.4; 30–64 years: OR, 1.8), to have not had surgery (OR, 2.2), had an Injury Severity Score (ISS) of 16 or lower (OR, 2.6), had a complication of care (OR, 4.4), and rated the hospital care as poor (OR, 9.2). CONCLUSION A trauma patient who is satisfied with his or her physician care is one who is 65 years or older, requires surgery, and is predominantly satisfied with other aspects of their hospital care. Unsatisfied patients are younger, are nonoperative, had lower ISS, had a complication of care, and rated their hospital care as poor. Understanding the specific characteristics of Press-Ganey results for trauma patients will allow trauma surgeons and their hospital partners to develop strategies to improve patients’ satisfaction with their trauma surgeon’s care. LEVEL OF EVIDENCE Epidemiologic study, level III; therapeutic study, level IV.

Has TRISS become an anachronism? A comparison of mortality between the National Trauma Data Bank and Major Trauma Outcome Study databases.

BACKGROUND The Trauma and Injury Severity Score (TRISS) has been the approach to trauma outcome prediction during the past 20 years and has been adopted by many commercial registries. Unfortunately, its survival predictions are based upon coefficients that were derived from a data set collected in the 1980s and updated only once using a data set collected in the early 1990s. We hypothesized that the improvements in trauma care during the past 20 years would lead to improved survival in a large database, thus making the TRISS biased. METHODS The TRISSs from the Pennsylvania statewide trauma registry (Collector, Digital Innovations) for the years 1990 to 2010. Observed-to-expected mortality ratios for each year of the study were calculated by taking the ratio of actual deaths (observed deaths, O) to the summation of the probability of mortality predicted by the TRISS taken over all patients (expected deaths, E). For reference, O/E ratio should approach 1 if the TRISS is well calibrated (i.e., has predictive accuracy). RESULTS There were 408,489 patients with complete data sufficient to calculate the TRISSs. There was a significant trend toward improved outcome (i.e., decreasing O/E ratio; nonparametric test of trend, p < 0.001) over time in both the total population and the blunt trauma subpopulation. In the penetrating trauma population, there was a trend toward improved outcome (decreasing O/E ratio), but it did not quite reach significance (nonparametric test of trend p = 0.073). CONCLUSION There is a steady trend toward improved O/E survival in the Pennsylvania database with each passing year, suggesting that the TRISS is drifting out of calibration. It is likely that improvements in care account for these changes. For the TRISS to remain an accurate outcome prediction model, new coefficients would need to be calculated periodically to keep up with trends in trauma care. This requirement for occasional updating is likely to be a requirement of any trauma prediction model, but because many other deficiencies in the TRISS have been reported, we think that rather than updating the TRISS, it would be more productive to replace the TRISS with a modern statistical model. LEVEL OF EVIDENCE Prognostic study, level II.

Determining venous thromboembolic risk assessment for patients with trauma: the Trauma Embolic Scoring System.

BACKGROUND This study aimed to determine the relative “weight” of risk factors known to be associated with venous thromboembolism (VTE) for patients with trauma based on injuries and comorbidities. METHODS A retrospective review of 16,608 consecutive admissions to a trauma center was performed. Patients were separated into those who developed VTE (n = 141) versus those who did not (16,467). Univariate analysis was performed for each risk factor reported in the trauma literature. Risk factors that were shown to be significant (p < 0.05) by univariate analysis underwent multivariate analysis to develop odds ratios for VTE. The Trauma Embolic Scoring System (TESS) was derived from the multivariate coefficients. The resulting TESS was compared with a data set from the National Trauma Data Bank (2002–2006) to determine its ability to predict VTE. RESULTS The multivariate analysis demonstrated that age, Injury Severity Score, obesity, ventilator use for more than 3 days, and lower-extremity trauma were significant predictors of VTE in our patient population. The TESS was from 0 to 14, with the best prediction for those patients with a score of more than 6 (sensitivity, 81.6%; specificity, 84%). Overall, the model had excellent discrimination in predicting VTE with a receiver operating characteristic curve of 0.89. The VTE rates for TESS in the National Trauma Data Bank data set were similar for all integers except for 3 and 4, in which the VTE rates were significantly higher (3, 0.2% vs. 0.6%; 4, 0.4% vs. 1.0%). CONCLUSION The TESS provides an objective measure of classifying VTE risk for patients with trauma. The TESS could allow informed decision making regarding prophylaxis strategies in patients with trauma. LEVEL OF EVIDENCE Prognostic study, level II.

Magnet hospitals are a magnet for higher survival rates at adult trauma centers.

BACKGROUND Little is known about nursing care’s impact on trauma outcomes. The Magnet Recognition Program recognizes hospitals for quality patient care and nursing excellence based on objective standards. We hypothesized that Magnet-designated trauma centers would have improved survival over their non-Magnet counterparts. METHODS All 2009 to 2011 admissions to Pennsylvania’s Level I and II trauma centers with more than 500 admissions during the study period (10 Magnet and 17 non-Magnet hospitals) were extracted from the Pennsylvania Trauma Systems Foundation State Registry. A logistic regression model with mortality as the dependent variable included the following variables: Magnet status, age, sex, admitting temperature, logit transformation of mortality probability predicted by the Trauma Mortality Prediction Model (TMPM-ais), systolic blood pressure, mechanism of injury, paralytic drug use, and Glasgow Coma Scale motor (GCSm) score. RESULTS A total of 73,830 patients from the Pennsylvania Trauma Outcome Study database met inclusion criteria for this study. The Magnet and non-Magnet hospital groups were statistically indistinguishable with respect to level of designation, medical school association, surgical residency programs, in-house surgeons, and urban locations. Patients admitted to a Magnet hospital had a significantly decreased odds of mortality when compared with their non-Magnet counterparts (odds ratio, 0.83; 95% confidence interval, 0.70–0.99; p = 0.033), when controlling for numerous factors. Overall, the model has outstanding discrimination with a receiver operating characteristic curve of 0.93. CONCLUSION Admission to a Magnet-designated hospital is associated with a 20% reduction in mortality. We believe that the Magnet program’s attention to nursing competence has important consequences for trauma patients, as reflected in the improved survival rates in trauma patients admitted to Magnet-designated hospitals. LEVEL OF EVIDENCE Epidemiologic/prognostic study, level III. Care management study, level IV.

Hyponatremia as a fall predictor in a geriatric trauma population

INTRODUCTION Approximately one in three older adults fall each year, resulting in a significant proportion of geriatric traumatic injuries. In a hospital with a focus on geriatric fall prevention, we sought to characterize this population to develop targeted interventions. As mild hyponatremia, defined as a serum sodium <135meq/L, has been reported to be associated with falls, unsteadiness and attention deficits, we hypothesized that hyponatremia is associated with falls in our geriatric trauma population. METHODS Gender, age, pre-existing conditions (cardiac disease, diabetes, hematologic disorder, liver disease, malignancy, musculoskeletal disorder, neurological disorder, obesity, psychiatric disorder, pulmonary disease, renal disease, thyroid disease), mechanism of injury and admitting serum sodium level were queried for all geriatric trauma admissions from 2008 to 2011. Mechanism of injury was coded as falls admissions and non-falls admissions. Admitting serum sodium levels were coded as hyponatremic (<135mmol/L) and not hyponatremic (≥135mmol/L). RESULTS Of the 2370 geriatric trauma admissions during the study period, there were 1841 (77.7%) falls admissions and 293 (12.4%) patients who were hyponatremic. Gender, age, neurological disorder, hematologic disorder, and hyponatremia were found to be significant predictors of falls in both univariate and multivariable analyses. CONCLUSION Hyponatremic patients are significantly more likely to be admitted for a fall than non-hyponatremic patients, when adjusting for age, neurological disorder, and hematologic disorder. Consequently, hyponatremia identification and management should be an integral part of any geriatric trauma fall prevention programme. Additionally, if hyponatremia is found during a geriatric fall workup, it should be corrected prior to discharge and closely monitored by a primary care physician to prevent recurrent episodes of falls.

Process factors affecting door to percutaneous coronary intervention for acute myocardial infarction patients.

The purpose of this cross-sectional study was to identify key predictor variables with the most impact on door-to-balloon time for acute myocardial infarction patients. The authors examined arrival, process, and patient-related variables from retrospective data from calendar years 2006 and 2007 within a single community hospital (N = 273). The door-to-balloon time ranged from 28 to 167 minutes, with an average of 76.77 (standard deviation ±24.5) minutes. Key predictor variables identified through multivariate linear regression included portable chest X-ray, presentation from walk-in versus ambulance, responding cardiology group, emergency department (ED) time of arrival (day 8 am to 5 pm or night 5 pm to 8 am), ED day of arrival (weekday or weekend), if a code R was called prior to arrival, and if the patient was identified as having chest pain on admission to the ED. For patients with acute myocardial infarction at a single study site, the authors identified a number of key factors that delay prompt reperfusion.

Early neurological deterioration in older adults with traumatic brain injury

INTRODUCTION Traumatic brain injuries (TBIs) and resulting fatalities among older adults increased considerably in recent years. Neurological deterioration often goes unrecognized at the injury scene and patients arrive at emergency departments with near-normal Glasgow Coma Scale (GCS) scores. This study examined the proportion of older adults experiencing early neurological deterioration (prehospital to emergency department), associated factors, and association of the magnitude of neurological deterioration with TBI severity. METHODS This secondary analysis of National Trauma Data Bank Research Datasets included patients who were age ⩾65, sustained a TBI, and transported from the injury scene to an emergency department. Data analysis included chi-square analysis, t-tests, and logistic regression. Long-term anticoagulant/antiplatelet therapy was not associated with deterioration. RESULTS Of the sample of 91,886 patients, 13,913 (15.1%) experienced early neurological deterioration. Adjusting for covariates, age, gender, head AISmax injury severity, and probability of death were associated with early deterioration. Patients with severe and critical head injuries had the highest odds of early neurological deterioration (OR=1.41 [CI=1.22-1.63] and OR=1.98 [CI=1.63-2.40], p<0.001). DISCUSSION/CONCLUSIONS Prehospital providers, nurses, physicians, and other providers have opportunities to optimize outcomes from older adult TBI through early recognition of neurological deterioration, rapid transport to facilities for definitive treatment, and targeted rehabilitation.

Mature Trauma Intensivist Model Improves Intensive Care Unit Efficiency But Not Mortality

Background: Although the Leap Frog intensivist staffing model has been shown to improve outcomes in the intensive care unit (ICU), to date, no one has examined the effect of an intensivist model in a dedicated trauma ICU. With stricter adherence to evidence-based protocols and 24-hour availability, we hypothesized that a mature intensivist model in a trauma ICU would decrease mortality. Methods: Level II trauma center trauma ICU admissions 2006 to 2011. The ICU care provided by 6 trauma intensivists. Two periods were compared: early (2006-2008) and mature (2009-2011). Patients matched on age, Injury Severity Score (ISS), preexisting conditions, and so on in a univariate analysis, with significant variables placed in a logistic regression model, with mortality as the outcome. Results: A total of 3527 patients (2999 excluding do not resuscitate status) were reviewed. Age ≥65 (odds ratio [OR] 2.38, P < .001), ISS ≥17 (OR 3.3, P < .001), coagulopathy (OR 1.64, P = .004), and anemia (OR 1.73, P = .02) were independent predictors of mortality. Multivariate logistic model encompassing these factors found no statistically significant differences in mortality across the 6-year period. The ICU efficiency showed significant improvements in terms of ventilator days (30.1% EARLY vs 24.4% MATURE; P < .001), decreases in mean consultant use per patient (0.55 ± 0.85 EARLY vs 0.40 ± 0.74 MATURE; P < .001), and increase in number of bedside procedures per patient (0.09 ± 0.48 EARLY vs 0.40 ± 0.74 MATURE; P < .001 Conclusions: Our mature intensivists staffing model shows improvement in ICU throughput (ventilator days, ICU days, decreased consultant use, and increased bedside procedures) but no survival benefit. Further improvements in overall trauma mortality may lie in the resuscitative and operative phase of patient care.