Errors reported in cross match laboratory: a prospective data analysis.

BACKGROUND

Human errors contribute to one half of all ABO-incompatible transfusions and transfusion-associated fatalities.

MATERIAL AND

METHODS

We report distribution, type and frequency of errors through a prospective study designed specifically to determine errors reported in the cross match lab with their clinical outcome, and to investigate the contributing factors, and underlying system problems.

RESULTS

A total of 342 errors (6.2 per 1000 samples) were reported with majority of the errors being clerical (87.1%) and occurred outside the blood bank (86.5%). Labelling errors were the most frequent incidents encountered with bedside being the major site of deviation. The rate of labeling errors was 6.4 errors per 1000 samples (0.64%) in 32,189 samples studied. Among 80,100 components transfused, the frequency of incorrect blood component transfusion (IBCT) was estimated to be 22.5/100,000 blood components transfused. Miscollected samples (WBIT) occurred at a rate of 1 in 1532 samples (0.65 per 1000 samples). More than half of these errors occurred during the day shift (9 errors per 1000 request form) but more with urgent demands (11 errors per 1000 request form).

CONCLUSION

This study indicates the importance of proper specimen labeling and implemented cost-effective, non-compromising policy of rejecting each mislabelled specimen and realises the importance of ongoing quality monitoring to improve laboratory performance.