doi:10.1016/j.jemermed.2009.05.016 Brief Reports CLINICAL SOBRIETY ASSESSMENT BY EMERGENCY PHYSICIANS IN BLUNT TRAUMA PATIENTS WITH ACUTE ALCOHOL EXPOSURE Simon A. Mahler, MD,* Sanjay Pattani, MD, MSHA,† Jesse Standifer, MSIII,* Gloria Caldito, PHD,* Steven A. Conrad, MD, PHD,* and Thomas C. Arnold, MD* *Department of Emergency Medicine, Louisiana State University Health Sciences Center, Shreveport, Louisiana and †Department of Emergency Medicine, University of Maryland, Baltimore, Maryland Reprint Address: Simon A. Mahler, MD, Department of Emergency Medicine, Louisiana State University Health Sciences Center-Shreveport, 1501 Kings Hwy, Shreveport, LA 71130 e Abstract—Background: Alcohol use increases injury risk and severity. However, few studies have evaluated the ability of emergency physicians (EPs) to accurately deter- mine sobriety. Objectives: To determine the predictive value of clinical sobriety assessment by EPs in blunt trauma patients with acute alcohol use. Materials and Methods: Blunt trauma patients, aged 18 – 65 years with suspected acute alcohol use, were prospectively enrolled in the study. EPs assessed study subjects before sample collection for blood alcohol level (BAL) and urine drug screen measure- ment. Alcohol exposure was considered significant if BAL was > 80 mg/dL. Sobriety (non-significant alcohol expo- sure) was defined as a BAL < 80 mg/dL. EP sobriety assessment was compared to measured BAL and predictive values were calculated. Agreement on significance of alco- hol exposure occurred if EP-estimated BAL > 80 mg/dL agreed with measured BAL > 80 mg/dL, or estimated BAL < 80 mg/dL agreed with measured BAL < 80 mg/dL. Chi- squared analysis was used to compare the proportion of correct physician assessments among patients with sobriety and those with significant alcohol exposure. Results: Of 158 enrolled subjects, 153 completed clinical assessment. EP assessment had a predictive value of 83% (95% confidence interval [CI] 77–90%) for significant alcohol exposure and 69% (95% CI 60 –78%) for sobriety. Agreement on the significance of alcohol exposure was 82% (125/153; 95% CI 76 – 88%). EPs identified 32% (11/34; 95% CI 17– 48%) of sober patients, but identified 96% (114/119; 95% CI 92– 99%) of patients with significant alcohol exposure. EP as- sessment was significantly less accurate in identifying sober patients (p < 0.01). Conclusions: Emergency physicians identified significant recent alcohol exposure in blunt trauma patients 96% of the time. However, clinical assess- ment by EPs in blunt trauma patients with recent alcohol use had only moderate predictive value for significant al- cohol exposure. Sober patients were frequently misidenti- fied as having significant alcohol exposure. © 2010 Elsevier Inc. e Keywords—alcohol; sobriety; blunt trauma INTRODUCTION The relationship between alcohol exposure and injury from falls, assaults, and motor vehicle collisions has been well documented in the literature, with multiple studies supporting the conclusion that alcohol substantially increases the risk of injury (1–3). In addition to increasing risk of injury, alcohol consumption is also associated with increased injury severity (4 –9). Although alcohol’s associ- ation with increased injury severity is likely multifacto- rial, missed injuries are probably responsible for some of this increase. It has been demonstrated that trauma pa- tients with alcohol exposure are more likely to have missed injuries than sober trauma patients (10). Blunt trauma patients with alcohol exposure may exhibit di- minished capacity to localize pain and to accurately RECEIVED: 4 November 2008; FINAL SUBMISSION RECEIVED: 2 April 2009; ACCEPTED: 8 May 2009 The Journal of Emergency Medicine, Vol. 39, No. 5, pp. 685– 690, 2010 Copyright © 2010 Elsevier Inc. Printed in the USA. All rights reserved 0736-4679/$–see front matter 685