Sources of Barcode Scanner Failure on POCT Devices

Abstract

Abstract Barcodes offer a means of reducing identification errors from manual data entry. However, during implementation of barcodes in our intensive care unit, staff continued to use manual data entry due to the failure of the scanners to read the barcoded operator identification (ID) and patient bands on the first attempt. This study investigated the sources of barcode scanner failure. Two point-of-care testing (POCT) devices were examined: the i-Stat1 (East Windsor, NJ) and the Abbott PCx glucose meter (Bedford, MA). Patient bands and operator IDs used the interleaved 2 of 5 barcode symbology. For this trial, patient bands were placed on a staff member and his teenage daughter for 11 days (ankle and wrist) and 9 days (ankle) respectively (about twice the average institutional length of patient stay [4.9 days]). These bands were worn during normal activity, exercise, and showers. The i-Stat1 was less successful at scanning the 3 bands (9.1–47.5% failure rate, n = 22–71 attempts) than the PCx meter (2.9–30.8% failure rate, n = 20–68 attempts; n = 4 operators), although this difference was only significant for 1 of the ankle bands between devices (P = 0.03) and for scanning the operator ID between operators (P = 0.04). To determine device differences better, the worn bands were compared with new bands at various distances, angles, lighting, and barcode orientations. Staff were significantly more successful at scanning the new bands and ID badges when compared with worn bands on both devices (P