Electronic patient blood management monitoring using routine health record data: A proof‐of‐principle study monitoring perioperative tranexamic acid use
Muhammad Naim Che Rahimi, Paolo Polzella, Mehir Amin, Gardash Bakhishli, Simon Godley, Michael F. Murphy, Simon J. Stanworth, Hayley G. Evans
Source abstract
Abstract Background Monitoring Patient Blood Management (PBM) practices against evidence‐based standards is essential for quality improvement; however, current approaches are limited. In the UK, perioperative tranexamic acid (TXA) use is a national quality standard, yet monitoring relies on manual audit cycles that are resource‐intensive and limited in scope. We evaluated whether an audit could be automated using routinely collected electronic health record (EHR) data. Methods We performed a retrospective study at a tertiary NHS center using linked perioperative and transfusion datasets. Automated compliance indicators were constructed using coded procedures (denominator) and digitally documented TXA administration from WHO Surgical Safety Checklists and electronic prescribing records (numerator). A structured validation framework assessed data extractability, completeness, denominator coverage, coding accuracy, and concordance between electronic sources. Outputs were assessed by specialty and procedure and compared with contemporaneous manual audit findings. Results Between July–September 2025, 800 eligible procedures were identified. Comparison with an independent dataset demonstrated procedural coverage of 96.2% and miscoding rate of 3.9%. Overall automated TXA compliance was 86.3%. Concordance between WHO checklist and electronic prescription was 74.2%, with explainable discordance patterns. Substantial inter‐specialty variation was identified, ranging from 98.2% (trauma and orthopedics) to 0% (vascular surgery). Compared with October–December 2024, overall compliance increased by 7.6%. Discussion Automated EHR‐based audit of perioperative TXA compliance is feasible and demonstrates good validity. Structured validation confirmed data reliability, and full‐population extraction revealed granular specialty‐ and procedure‐level variation, likely undetectable by manual audits, supporting its wider evaluation as a continuous PBM quality monitoring tool.
Evidence graph
No public relationships recorded yet.
Integrity note: This page is a factual metadata record created by deterministic ingestion. It is not a claim that the work moves a mathematical frontier or has been independently verified.