ABSTRACT
BACKGROUND: Abstracting cerebrovascular disease (CeVD) from inpatient electronic medical records (EMRs) through natural language processing (NLP) is pivotal for automated disease surveillance and improving patient outcomes. Existing methods rely on coders' abstraction, which has time delays and under-coding issues. This study sought to develop an NLP-based method to detect CeVD using EMR clinical notes. METHODS: CeVD status was confirmed through a chart review on randomly selected hospitalized patients who were 18 years or older and discharged from 3 hospitals in Calgary, Alberta, Canada, between January 1 and June 30, 2015. These patients' chart data were linked to administrative discharge abstract database (DAD) and Sunrise™ Clinical Manager (SCM) EMR database records by Personal Health Number (a unique lifetime identifier) and admission date. We trained multiple natural language processing (NLP) predictive models by combining two clinical concept extraction methods and two supervised machine learning (ML) methods: random forest and XGBoost. Using chart review as the reference standard, we compared the model performances with those of the commonly applied International Classification of Diseases (ICD-10-CA) codes, on the metrics of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). RESULT: Of the study sample (n = 3036), the prevalence of CeVD was 11.8% (n = 360); the median patient age was 63; and females accounted for 50.3% (n = 1528) based on chart data. Among 49 extracted clinical documents from the EMR, four document types were identified as the most influential text sources for identifying CeVD disease ("nursing transfer report," "discharge summary," "nursing notes," and "inpatient consultation."). The best performing NLP model was XGBoost, combining the Unified Medical Language System concepts extracted by cTAKES (e.g., top-ranked concepts, "Cerebrovascular accident" and "Transient ischemic attack"), and the term frequency-inverse document frequency vectorizer. Compared with ICD codes, the model achieved higher validity overall, such as sensitivity (25.0% vs 70.0%), specificity (99.3% vs 99.1%), PPV (82.6 vs. 87.8%), and NPV (90.8% vs 97.1%). CONCLUSION: The NLP algorithm developed in this study performed better than the ICD code algorithm in detecting CeVD. The NLP models could result in an automated EMR tool for identifying CeVD cases and be applied for future studies such as surveillance, and longitudinal studies.
ABSTRACT
BACKGROUND: Intraventricular hemorrhage requiring ventriculostomy placement is a frequent complication of spontaneous intracerebral hemorrhage. Although a subset of patients will require permanent ventricular shunt placement, little is known about contemporary practices regarding the timing of ventriculostomy and ventricular shunt placement after intracerebral hemorrhage. METHODS: Using the 2010-2012 National Inpatient Sample, we identified patients with International Classification of Diseases, Ninth Revision codes for intracerebral hemorrhage, excluded secondary causes, and examined procedure dates. RESULTS: Of 35,899 patients with primary intracerebral hemorrhage, 2443 (6.8%) received ventriculostomy, 93% within the first 3 days of admission and 66% within the first day. Permanent shunt placement occurred in 173 (7.1%) patients following ventriculostomy at a median interval of 15 days (interquartile range: 11-20). Among those remaining alive and in hospital at 14, 21, and 28 days, 5%, 11%, and 15%, respectively, underwent shunt placement following ventriculostomy, and 24% of those with multiple ventriculostomy insertions required permanent shunt by 4 weeks of hospitalization. Multiple ventriculostomies, tracheostomy, and black race were associated with longer time to permanent shunt. CONCLUSIONS: A wide variation in delay to permanent shunt placement is present, with substantial and increasing prevalence with time in hospital. Better understanding of the risk factors associated with persistent hydrocephalus will help optimize patient selection and timing of treatment.