Barcode error leading to sample misidentification during blood grouping

Patient safety associated with the transfusion process is extremely important due to the propensity of errors to cause catastrophic consequences. To aid blood banks in minimizing these errors, advanced computerized information systems and barcode labeling technology have been introduced. The barcode technology comprises machine readable symbols used to encode information to automate it. It simplifies and improves the patient identification system in laboratory and clinical transfusion practice.[1] Currently barcode technology has become an indispensable advancement allowing the technical staff to bypass the tedious manual check of the sample labels.

Our center is a 2032-bed tertiary care facility and administers approximately 30,000 blood components annually. We have a state-of-the-art immunohematology laboratory with fully automated grouping system with an annual load of 54,000 patient samples. Samples are barcoded at the site of reception at the blood bank, using a barcode printer of Zebra Technologies Corporation. Subsequently samples are loaded into automated blood grouping equipment, which has a software into which the patient details are fed manually and the labels are scanned by an built-in scanner. Recently we noticed a blood grouping error (O D+ instead of B D+) and on root-cause analysis it was found to be due to barcode printing error.

As visualized in Video Clip S1 (available as supporting information in the online version of this paper), the barcode label printed in the above said barcode printer in our blood bank, for a particular hospital number, was read as a different number belonging to another patient, leading to the blood grouping error. While entering the patient's blood grouping report to the blood bank software, a discrepancy was noted with the previous grouping report. To resolve the discrepancy, we sought a fresh sample from the patient. Error due to wrong blood in tube was ruled out by repeat blood grouping with a fresh sample. Subsequently, to preclude the possibility of a technical error by the automated grouping instrument we checked the event log in the system, which showed the loading of duplicate sample of the patient and as per system protocol, it accepted the first sample. On visual inspection, we could not trace out any duplicate sample, as there was no error in keystroke entry of the patient name or the hospital number. However, we identified the mistake while scanning the barcode label on the samples. Barcode label error can occur due to various reasons as depicted in Fig. 1.[2, 3] In our root-cause analysis, we identified a barcode printing error due to printer resolution defect as the reason underlying the error shown in Video Clip S1. The major disadvantage with linear barcodes is that they only contain one identifier. Hence, it will be ideal if health care providers follow a stringent quality control program for the barcode printers or adopt a higher-fidelity alternative like two-dimensional barcodes for patient safety.