RESUMEN
Background / Objective: Clinical documentation is essential for conveying medical decision-making, communication between providers and patients, and capturing quality, billing, and regulatory measures during emergency department (ED) visits. Growing evidence suggests the benefits of note template standardization, however, variations in documentation practices are common. The primary objective of this study is to measure the utilization and coding performance of a standardized ED note template implemented across a nine-hospital health system. METHODS: This was a retrospective study before and after the implementation of a standardized ED note template. A multi-disciplinary group consensus was built around standardized note elements, provider note workflows within the electronic health record (EHR), and how to incorporate newly required medical decision-making elements. The primary outcomes measured included the proportion of ED visits using standardized note templates, and the distribution of billing codes in the six months before and after implementation. RESULTS: In the pre-implementation period, a total of six legacy ED note templates were being used across nine emergency departments, with the most used template accounting for approximately 36% of ED visits. Marked variations in documentation elements were noted across six legacy templates. After the implementation, 82% of ED visits system-wide used a single standardized note template. Following implementation, we observed a 1% increase in the proportion of ED visits coded as highest acuity and an unchanged proportion coded as second highest acuity. CONCLUSIONS: We observed a greater than two-fold increase in the use of a standardized ED note template across a 9-hospital health system in anticipation of the new 2023 coding guidelines. The development and utilization of a standardized note template format relied heavily on multi-disciplinary stakeholder engagement to inform design that worked for varied documentation practices within the EHR. After the implementation of a standardized note template, we observed better-than-anticipated coding performance.
RESUMEN
STUDY OBJECTIVE: Patients undergoing diagnostic imaging studies in the emergency department (ED) commonly have incidental findings, which may represent unrecognized serious medical conditions, including cancer. Recognition of incidental findings frequently relies on manual review of textual radiology reports and can be overlooked in a busy clinical environment. Our study aimed to develop and validate a supervised machine learning model using natural language processing to automate the recognition of incidental findings in radiology reports of patients discharged from the ED. METHODS: We performed a retrospective analysis of computed tomography (CT) reports from trauma patients discharged home across an integrated health system in 2019. Two independent annotators manually labeled CT reports for the presence of an incidental finding as a reference standard. We used regular expressions to derive and validate a random forest model using open-source and machine learning software. Final model performance was assessed across different ED types. RESULTS: The study CT reports were divided into derivation (690 reports) and validation (282 reports) sets, with a prevalence of incidental findings of 22.3%, and 22.7%, respectively. The random forest model had an area under the curve of 0.88 (95% confidence interval [CI], 0.84 to 0.92) on the derivation set and 0.92 (95% CI, 0.88 to 0.96) on the validation set. The final model was found to have a sensitivity of 92.2%, a specificity of 79.4%, and a negative predictive value of 97.2%. Similarly, strong model performance was found when stratified to a dedicated trauma center, high-volume, and low-volume community EDs. CONCLUSION: Machine learning and natural language processing can classify incidental findings in CT reports of ED patients with high sensitivity and high negative predictive value across a broad range of ED settings. These findings suggest the utility of natural language processing in automating the review of free-text reports to identify incidental findings and may facilitate interventions to improve timely follow-up.