Detection of allergic transfusion‐related adverse events from electronic medical records

Abstract

Background Transfusion‐related adverse events can be unrecognized and unreported. As part of the US Food and Drug Administration's Center for Biologics Evaluation and Research Biologics Effectiveness and Safety initiative, we explored whether machine learning methods, such as natural language processing (NLP), can identify and report transfusion allergic reactions (ARs) from electronic health records (EHRs). Study Design and Methods In a 4‐year period, all 146 reported transfusion ARs were pulled from a database of 86,764 transfusions in an academic health system, along with a random sample of 605 transfusions without reported ARs. Structured and unstructured EHR data were retrieved, including demographics, new symptoms, medications, and lab results. In unstructured data, evidence from clinicians' notes, test results, and prescriptions fields identified transfusion ARs, which were used to extract NLP features. Clinician reviews of selected validation cases assessed and confirmed model performance. Results Clinician reviews of selected validation cases yielded a sensitivity of 67.9% and a specificity of 97.5% at a threshold of 0.9, with a positive predictive value (PPV) of 84%, estimated to 4.5% when extrapolated to match transfusion AR incidence in the full transfusion dataset. A higher threshold achieved sensitivity of 43% with specificity/PPV of 100% in our validation set. Essential features predicting ARs were recognized transfusion reactions, administration of antihistamines or glucocorticoids, and skin symptoms (e.g., hives and itching). Removal of NLP features decreased model performance. Discussion NLP algorithms can identify transfusion reactions from the EHR with a reasonable level of precision for subsequent clinician review and confirmation.

Machine learning in transfusion medicine: A scoping review

“…ML has been applied primarily to enhance the ability to detect and predict acute transfusion reactions (ATRs) and adverse transfusion events. Novel information retrieval methods, such as natural language processing (NLP), when applied to electronic health records (EHRs) have demonstrated underreporting by clinicians and the potential to improve detection 80–82 . Alternatively, Roubinian et al 83 and Nguyen et al 84 incorporated novel biomarkers into classification models and decision tree analysis, respectively.…”

“…Novel information retrieval methods, such as natural language processing (NLP), when applied to electronic health records (EHRs) have demonstrated underreporting by clinicians and the potential to improve detection. [80][81][82] Alternatively, Roubinian et al 83 and Nguyen et al 84 incorporated novel biomarkers into classification models and decision tree analysis, respectively. While the focus of ML is on prediction, and a causal relationship cannot be assumed of the covariates found to have high predictive value, identification of novel risk factors for hypothesis generation and further research can be useful as seen in transfusion-associated lung injury (TRALI) 85 and in pediatric transfusion-associated hyperkalemia.…”

The incidence of transfusion‐related acute lung injury using active surveillance: A systematic review and meta‐analysis

“…[14][15][16]94 Use of novel data-driven approaches to active surveillance of electronic health records, such as large language models or other machine learning methods could facilitate enhanced detection. 100 In conclusion, we used meta-analysis to obtain estimates for TRALI rates for RBCs, platelets, and plasma, by component transfused. This information is important when considering the risks and benefits of blood transfusion, which are essential elements to informed consent and transfusion decisions.…”

“…This could take the form of broader implementation of active surveillance; however, in practice, active surveillance can be highly resource intensive, which presents challenges outside of a research setting 14–16,94 . Use of novel data‐driven approaches to active surveillance of electronic health records, such as large language models or other machine learning methods could facilitate enhanced detection 100 …”

Machine learning applications and challenges in graft-versus-host disease: a scoping review

“…In transfusion medicine, which is an essential component of HCTs, ML can enhance the safety of blood products by mitigating transfusion-related adverse events [34,35]. For instance, natural language processing, an ML method that was trained on electronic health records, can identify patients at risk of allergic reactions to transfusions or predict the need for perioperative red blood cell transfusion [36]. In the field of hematology, ML can categorize acute leukemia subtypes using flow cytometric data [37], predict the prognosis of multiple myeloma based on gene expression profiles [38], or identify novel biomarkers for myelodysplastic syndromes using DNA methylation patterns [39].…”