Prediction of intra-abdominal injury using natural language processing of electronic medical record data.

BACKGROUND: This study aimed to use natural language processing to predict the presence of intra-abdominal injury using unstructured data from electronic medical records. METHODS: This was a random-sample retrospective observational cohort study leveraging unstructured data from injured patients taken to one of 9 acute care hospitals in an integrated health system between 2015 and 2021. Patients with International Classification of Diseases External Cause of Morbidity codes were identified. History and physical, consult, progress, and radiology report text from the first 8 hours of care were abstracted. Annotator dyads independently annotated encounters' text files to establish ground truth regarding whether intra-abdominal injury occurred. Features were extracted from text using natural language processing techniques, bag of words, and principal component analysis. We tested logistic regression, random forests, and gradient boosting machine to determine accuracy, recall, and precision of natural language processing to predict intra-abdominal injury. RESULTS: A random sample of 7,000 patient encounters of 177,127 was annotated. Only 2,951 had sufficient information to determine whether an intra-abdominal injury was present. Among those, 84 (2.9%) had an intra-abdominal injury. The concordance between annotators was 0.989. Logistic regression of features identified with bag of words and principal component analysis had the best predictive ability, with an area under the receiver operating characteristic curve of 0.9, recall of 0.73, and precision of 0.17. Text features with greatest importance included "abdomen," "pelvis," "spleen," and "hematoma." CONCLUSION: Natural language processing could be a screening decision support tool, which, if paired with human clinical assessment, can maximize precision of intra-abdominal injury identification.

The gap between hospital-based violence intervention services and client needs: A systematic review.

BACKGROUND: Survivors of intentional interpersonal violence face social challenges related to social determinants of health that led to their initial injury. Hospital-based violence intervention programs reduce reinjury. It is unclear how well they meet clients' reported needs. This systematic review aimed to quantify how well hospital-based violence intervention program services addressed clients' reported needs. METHODS: Medline, The Cochrane Library, CINAHL Plus with Full Text, and PsycInfo were queried for studies addressing hospital-based violence intervention programs services and intentional injury survivors' needs in the United States. Case reports, reviews, editorials, theses, and studies focusing on pediatric patients, victims of intimate partner violence, or sexual assault were excluded. Data extracted included program structure, hospital-based violence intervention program services, and client needs assessments before and after receiving hospital-based violence intervention program services. RESULTS: Of the 3,339 citations identified, 13 articles were selected for inclusion. Hospital-based violence intervention programs clients' most reported needs included mental health (10 studies), employment (7), and education (5) before receiving hospital-based violence intervention programs services. Only 4 studies conducted quantitative client needs assessments before and after receiving hospital-based violence intervention program services. All 4 studies were able to meet at least 50% of each of the clients' reported needs. The success rate depended on the need and program location: success in meeting mental health needs ranged from 65% to 90% of clients. Conversely, time-intensive long-term needs were least met, including employment 60% to 86% of clients, education 47% to 73%, and housing 50% to 71%. CONCLUSION: Few hospital-based violence intervention programs studies considered clients' reported needs. Employment, education, and housing must be a stronger focus of hospital-based violence intervention programs.