Human errors in manual techniques for ABO/D grouping are associated with potentially lethal outcomes

AIMS/OBJECTIVES: To review if ABO/D grouping errors are more likely to occur with manual intervention compared to automation. BACKGROUND: Human errors in manual pre-transfusion testing may result in ABO/D-incompatible transfusions and catastrophic outcomes. Accurate ABO/D grouping is a critical part of pre-transfusion testing. METHODS: This was a retrospective analysis of reports made to Serious Hazards of Transfusion (SHOT) between January 2004 and December 2016 where ABO/D grouping errors led to the transfusion of an incorrect blood component to review if errors are more likely to occur with manual intervention compared to automation. RESULTS: In 148 of 158 (93%) ABO/D grouping errors, manual intervention took place. In the remaining 10, causes were not reported. No errors occurred with full automation. Interpretation errors occurred in 86 of 148 (58%) and 42 of 148 (28%) transcription errors, and in 20 of 148, wrong or no samples were selected. Of 148 errors, 21 (14%) resulted in ABO-incompatible transfusion, with one death in 2004 due to an interpretation error in a manual ABO group. In 30 of 148 (20%), D-positive red cells were given to D-negative recipients, where three women of child-bearing potential became sensitised and developed anti-D. ABO grouping errors have reduced from 18 of 539 (3%) of total reports analysed in 2004 (3·3%) to 3 of 3091 (0·10%) in 2016. CONCLUSIONS: Where manual testing cannot be avoided, results should be confirmed using automated techniques as soon as possible, and a back-up process should be available 24/7. SHOT data confirm that manual interventions are prone to human error, especially in transcription and interpretation, and demonstrate a continuing need for appropriate serological knowledge and understanding by transfusion laboratory staff to underpin safety provided by automation and information technology (IT).