Clinical validation of a delta check model in haematology automated counting improves data validation.

INTRODUCTION: The delta check quality control measure has been applied in haematology testing for a long time. Although the delta check graph of haematology test had been reported by Berend Houwen in 1989, problems still exist in its clinical application. This study aims to provide a simplified and practical process for the improvement of validity in haematology testing. METHODS: We analysed a total of 5260 patients with two sequential complete blood counting results, taken in 1 week, to verify the delta check curve. The delta check equations were simulated using the sigma plot 13.0 software. Validations of test results were performed on patients from the haematology, oncology and other departments. RESULTS: Three equations of the delta check limits for mean corpuscular volume (MCV), haemoglobin, white blood cell and platelets were simulated. The following two limitation lines were selected: limit A (ΔA) and limit B (ΔB), wherein ΔA was to simplify the review rules and ΔB to check random errors. Delta check procedures including rules and equations were established for clinical use, and the re-inspection rate was reduced in different departments using the delta check model in clinical validation. CONCLUSION: The simplified delta check limitation formulae are meaningful for data review and may help the laboratory technician optimize workflow and reduce workload.