Coagulation and Transfusion Informatics in Chronic Liver Disease: A Data Linkage Study of Emergency Department Presentations.

OBJECTIVE

To describe the application of data linkage for analysing coagulation abnormalities and blood transfusion practices in patients with chronic liver disease (CLD) presenting to emergency departments (EDs).

METHODS

Patients with CLD presenting to 104 Queensland Health EDs (January 2016-August 2023) were identified using International Classification of Diseases codes. Phase 1 deterministically linked ED, admission, pathology, transfusion and death records using unique identifiers. Phase 2 used Structured Query Language (SQL) to capture transfusion timing. The model incorporated data from both digital and non-digitalised hospitals.

RESULTS

Phase 1 linkage identified 36,643 ED presentations, 443,367 admissions, 47,357 deaths, 3,004,236 pathology results and 140,687 blood transfusion events. Phase 2 identified 23,578 ED presentations by 11,961 patients, linked to 20,312 admissions, 921 deaths, 22,284 full blood counts (FBC), 19,408 coagulation profiles, 3068 blood gases, 457 rotational thromboelastometry (ROTEM) and 53 thromboelastography tests. Transfusion data were linked to 1616 presentations, including of 1358 red blood cell (RBC) transfusion episodes, 330 fresh frozen plasma, 324 cryoprecipitate, 418 platelets, 51 fibrinogen concentrate and 280 Prothrombinex-VF administration episodes. High linkage rates were achieved for FBC (99.4%), coagulation profile (97.6%) and biochemistry (92.3%), while linkages for blood gas (34.6%), ROTEM (13.8%) and thromboelastography (2.5%) were less frequent. Massive transfusions occurred in 27 presentations (≥ 4 RBC units in 4 h) and in 22 presentations (≥ 10 RBC units in 24 h), with 100% linkage for FBC and coagulation profiles in both groups.

CONCLUSION

The feasibility of data linkage to investigate coagulation abnormalities and transfusion in CLD patients was demonstrated. This model provides a scalable method for haemovigilance and transfusion research.

TRIAL REGISTRATION

The authors have confirmed clinical trial registration is not needed for this submission.