Natural language processing and machine learning to identify alcohol misuse from the electronic health record in trauma patients: development and internal validation.

Objective: Alcohol misuse is present in over a quarter of trauma patients. Information in the clinical notes of the electronic health record of trauma patients may be used for phenotyping tasks with natural language processing (NLP) and supervised machine learning. The objective of this study is to train and validate an NLP classifier for identifying patients with alcohol misuse. Materials and Methods: An observational cohort of 1422 adult patients admitted to a trauma center between April 2013 and November 2016. Linguistic processing of clinical notes was performed using the clinical Text Analysis and Knowledge Extraction System. The primary analysis was the binary classification of alcohol misuse. The Alcohol Use Disorders Identification Test served as the reference standard. Results: The data corpus comprised 91 045 electronic health record notes and 16 091 features. In the final machine learning classifier, 16 features were selected from the first 24 hours of notes for identifying alcohol misuse. The classifier's performance in the validation cohort had an area under the receiver-operating characteristic curve of 0.78 (95% confidence interval [CI], 0.72 to 0.85). Sensitivity and specificity were at 56.0% (95% CI, 44.1% to 68.0%) and 88.9% (95% CI, 84.4% to 92.8%). The Hosmer-Lemeshow goodness-of-fit test demonstrates the classifier fits the data well (P = .17). A simpler rule-based keyword approach had a decrease in sensitivity when compared with the NLP classifier from 56.0% to 18.2%. Conclusions: The NLP classifier has adequate predictive validity for identifying alcohol misuse in trauma centers. External validation is needed before its application to augment screening.

Screening for at-risk drinking behavior in trauma patients.

CONTEXT: A blood alcohol level above 0 g/dL is found in up to 50% of patients presenting with traumatic injuries. The presence of alcohol in the blood not only increases the risk of traumatic injury, but it is also associated with worse outcomes and trauma recidivism. In light of these risks, the American College of Surgeons Committee on Trauma advocates screening for at-risk drinking. Although many institutions use blood alcohol levels to determine at-risk drinking in trauma patients, the Alcohol Use Disorders Identification Test (AUDIT) offers a cheap and easy alternative. Few direct comparisons have been made between these 2 tests in trauma patients. OBJECTIVE: To compare the utility of blood alcohol level and AUDIT score as indicators of at-risk drinking in trauma patients. METHODS: Records for all trauma patients aged 18 years or older who were admitted to a level I trauma center from May 2013 through June 2014 were reviewed in this retrospective cohort study. Inclusion criteria required patients to have undergone both blood alcohol level testing and AUDIT on admission. A blood alcohol level greater than 0 g/dL and an AUDIT score equal to or above 8 were considered positive for at-risk drinking. Performance of both tests was indexed against the National Institute of Alcohol Abuse and Alcoholism (NIAAA) criteria for at-risk drinking. RESULTS: Of 750 patients admitted for trauma, 222 records (30%) contained data on both blood alcohol level and AUDIT score. The patients were predominantly male (178 [80%]) and had a mean (SD) age of 40.1 (16.7) years. Most patients (178 [80%]) had sustained blunt trauma. Ninety-seven patients (44%) had a positive blood alcohol level, 70 (35%) had a positive AUDIT score, and 54 (24%) met NIAAA criteria for at-risk drinking. The sensitivity and specificity of having a positive blood alcohol level identify at-risk drinking were 61% and 62%, respectively. The sensitivity and specificity of having a positive AUDIT score identify at-risk drinking were 83% and 81%, respectively. CONCLUSION: As a stand-alone indicator of at-risk drinking behavior in trauma patients, the AUDIT score was shown to be superior to blood alcohol level. The utility of obtaining routine blood alcohol levels in trauma patients as a screening tool for at-risk drinking should be reexamined.