Natural language processing of prehospital emergency medical services trauma records allows for automated characterization of treatment appropriateness.

BACKGROUND: Incomplete prehospital trauma care is a significant contributor to preventable deaths. Current databases lack timelines easily constructible of clinical events. Temporal associations and procedural indications are critical to characterize treatment appropriateness. Natural language processing (NLP) methods present a novel approach to bridge this gap. We sought to evaluate the efficacy of a novel and automated NLP pipeline to determine treatment appropriateness from a sample of prehospital EMS motor vehicle crash records. METHODS: A total of 142 records were used to extract airway procedures, intraosseous/intravenous access, packed red blood cell transfusion, crystalloid bolus, chest compression system, tranexamic acid bolus, and needle decompression. Reports were processed using four clinical NLP systems and augmented via a word2phrase method leveraging a large integrated health system clinical note repository to identify terms semantically similar with treatment indications. Indications were matched with treatments and categorized as indicated, missed (indicated but not performed), or nonindicated. Automated results were then compared with manual review, and precision and recall were calculated for each treatment determination. RESULTS: Natural language processing identified 184 treatments. Automated timeline summarization was completed for all patients. Treatments were characterized as indicated in a subset of cases including the following: 69% (18 of 26 patients) for airway, 54.5% (6 of 11 patients) for intraosseous access, 11.1% (1 of 9 patients) for needle decompression, 55.6% (10 of 18 patients) for tranexamic acid, 60% (9 of 15 patients) for packed red blood cell, 12.9% (4 of 31 patients) for crystalloid bolus, and 60% (3 of 5 patients) for chest compression system. The most commonly nonindicated treatment was crystalloid bolus (22 of 142 patients). Overall, the automated NLP system performed with high precision and recall with over 70% of comparisons achieving precision and recall of greater than 80%. CONCLUSION: Natural language processing methodologies show promise for enabling automated extraction of procedural indication data and timeline summarization. Future directions should focus on optimizing and expanding these techniques to scale and facilitate broader trauma care performance monitoring. LEVEL OF EVIDENCE: Diagnostic tests or criteria, level III.

Electronic Progress Note Reading Patterns: An Eye Tracking Analysis.

This study used eye-tracking to understand how the order of note sections influences the way physicians read electronic progress notes. Participants (n = 7) wore an eye-tracking device while reviewing progress notes for four patient cases and then provided a verbal summary. We reviewed and analyzed verbal summaries and eye tracking recordings. Wide variation in reading behaviors existed. There was no relationship between time spent reading a section and section origin of verbal summaries.

Recurrent Deep Network Models for Clinical NLP Tasks: Use Case with Sentence Boundary Disambiguation.

Although a number of foundational natural language processing (NLP) tasks like text segmentation are considered a simple problem in the general English domain dominated by well-formed text, complexities of clinical documentation lead to poor performance of existing solutions designed for the general English domain. We present an alternative solution that relies on a convolutional neural network layer followed by a bidirectional long short-term memory layer (CNN-Bi-LSTM) for the task of sentence boundary disambiguation and describe an ensemble approach for domain adaptation using two training corpora. Implementations using the Keras neural-networks API are available at https://github.com/NLPIE/clinical-sentences.

Named Entity Recognition in Prehospital Trauma Care.

Natural language processing (NLP) methods would improve outcomes in the area of prehospital Emergency Medical Services (EMS) data collection and abstraction. This study evaluated off-the-shelf solutions for automating labelling of clinically relevant data from EMS reports. A qualitative approach for choosing the best possible ensemble of pretrained NLP systems was developed and validated along with a feature using word embeddings to test phrase synonymy. The ensemble showed increased performance over individual systems.

Characterizing Functional Health Status of Surgical Patients in Clinical Notes.

Functional health status is an important factor not only for determining overall health, but also for measuring risks of adverse events. Our hypothesis is that important functional status data is contained in clinical notes. We found that several categories of phrases related to functional status including diagnoses, activity and care assessments, physical exam, functional scores, assistive equipment, symptoms, and surgical history were important factors. Use of functional health status level terms from our chart review compared to National Surgical Quality Improvement Program determination had varying sensitivities for correct functional status category identification, with 96% for independent patients, 60% for partially dependent patients, and 44% for totally dependent patients. Inter-rater agreement assessing term relevance to functional health status was high at 91% (Kappa=0.74). Functional status-related terms in clinical notes show potential for use in future methodologies for automated detection of functional health status for quality improvement registries and other clinical assessments.

Comparing Existing Resources to Represent Dietary Supplements.

Dietary supplements, often considered as food, are widely consumed despite of limited knowledge around their safety/efficacy and any well-established regulatory policies, unlike their drug counterparts. Informatics methods may be useful in filling this knowledge gap, however, the lack of standardized representation of DS hinders this progress. In this pilot study, five electronic DS resources, i.e., NM, DSID & NHPID (ingredient level) and DSLD & LNHPD (product level), were evaluated and compared both quantitatively and qualitatively employing four phases. Essential data elements needed for comprehensive DS representation were compiled based on LanguaL code (food) & AHFSA (drugs) guidelines and employed as a check-list. We further investigated the completeness of DS representation by incorporating Ginseng and Fish oil as examples. We found fragmented and inconsistent distribution of DS representation in terms of essential data elements across five resources. This study provides a preliminary platform for development of standardized DS terminology/ontology model.

Representation of Social History Factors Across Age Groups: A Topic Analysis of Free-Text Social Documentation.

As individuals age, there is potential for dramatic changes in the social and behavioral determinants that affect health status and outcomes. The importance of these determinants has been increasingly recognized in clinical decision-making. We sought to characterize how social and behavioral health determinants vary in different demographic groups using a previously established schema of 28 social history types through both manual analysis and automated topic analysis of social documentation in the electronic health record across the population of an entire integrated healthcare system. Our manual analysis generated 8,335 annotations over 1,400 documents, representing 24 (86%) social history types. In contrast, automated topic analysis generated 22 (79%) social history types. A comparative evaluation demonstrated both similarities and differences in coverage between the manual and topic analyses. Our findings validate the widespread nature of social and behavioral determinants that affect health status over populations of individuals over their lifespan.

Assessing the Representation of Occupation Information in Free-Text Clinical Documents Across Multiple Sources.

There has been increasing recognition of the key role of social determinants like occupation on health. Given the relatively poor understanding of occupation information in electronic health records (EHRs), we sought to characterize occupation information within free-text clinical document sources. From six distinct clinical sources, 868 total occupation-related sentences were identified for the study corpus. Building off approaches from previous studies, refined annotation guidelines were created using the National Institute for Occupational Safety and Health Occupational Data for Health data model with elements added to increase granularity. Our corpus generated 2,005 total annotations representing 39 of 41 entity types from the enhanced data model. Highest frequency entities were: Occupation Description (17.7%); Employment Status - Not Specified (12.5%); Employer Name (11.0%); Subject (9.8%); Industry Description (6.2%). Our findings support the value of standardizing entry of EHR occupation information to improve data quality for improved patient care and secondary uses of this information.