Pagali, Sandeep

Abstract Background Delirium is underdiagnosed in clinical practice and is not routinely coded for billing. Manual chart review can be used to identify the occurrence of delirium; however, it is labor-intensive and impractical for large-scale studies. Natural language processing (NLP) has the capability to process raw text in electronic health records (EHRs) and determine the meaning of the information. We developed and validated NLP algorithms to automatically identify the occurrence of delirium from EHRs. Methods This study used a randomly selected cohort from the population-based Mayo Clinic Biobank (N = 300, age ≥65). We adopted the standardized evidence-based framework confusion assessment method (CAM) to develop and evaluate NLP algorithms to identify the occurrence of delirium using clinical notes in EHRs. Two NLP algorithms were developed based on CAM criteria: one based on the original CAM (NLP-CAM; delirium vs no delirium) and another based on our modified CAM (NLP-mCAM; definite, possible, and no delirium). The sensitivity, specificity, and accuracy were used for concordance in delirium status between NLP algorithms and manual chart review as the gold standard. The prevalence of delirium cases was examined using International Classification of Diseases, 9th Revision (ICD-9), NLP-CAM, and NLP-mCAM. Results NLP-CAM demonstrated a sensitivity, specificity, and accuracy of 0.919, 1.000, and 0.967, respectively. NLP-mCAM demonstrated sensitivity, specificity, and accuracy of 0.827, 0.913, and 0.827, respectively. The prevalence analysis of delirium showed that the NLP-CAM algorithm identified 12 651 (9.4%) delirium patients, the NLP-mCAM algorithm identified 20 611 (15.3%) definite delirium cases, and 10 762 (8.0%) possible cases. Conclusions NLP algorithms based on the standardized evidence-based CAM framework demonstrated high performance in delineating delirium status in an expeditious and cost-effective manner.

BACKGROUND/OBJECTVES Four direct‐acting oral anticoagulants (DOACs) are currently approved by the Food and Drug Administration for the treatment of venous thromboembolism (VTE). Limited efficacy and safety data are available for their use in older adults (aged ≥75 years). METHODS Medline, Cochrane Central Register of Controlled Trials, Embase, EBSCO, Web of Science, and CINAHL databases were searched for trials comparing DOACs with vitamin K antagonists (VKAs) for the treatment of VTE in older adults from inception through January 1, 2020. Meta‐analysis was performed to assess the combined endpoint of recurrent VTE and related deaths and bleeding events (composite of major and clinically relevant nonmajor bleeding). The Mantel‐Haenszel relative risk (RR) random effects model was used to pool results across studies. RESULTS Six randomized controlled trials at low risk of bias met criteria for inclusion with a total of 3,665 patients aged 75 years and older with follow‐up of 24 weeks or longer. Data for bleeding events were not available for dabigatran. Overall, DOACs had an improved efficacy over VKAs (RR = .56; 95% confidence interval [CI] = .38‐.82). There was no statistically significant difference in the safety outcomes (RR = .77; 95% CI = .56‐1.05). No significant heterogeneity was observed for efficacy outcome, and only moderate heterogeneity was observed for safety outcome. CONCLUSION In older adults with VTE, DOACs appear to improve rates of recurrent VTE and VTE‐related deaths compared with VKAs with similar bleeding outcomes.