RESUMEN
OBJECTIVE: To evaluate the effect of the incorporation of the semi-automated Kardex medication storage and dispensing system on the quality of the filling process of medicine trolleys in unit doses. METHOD: The speed of filling medicine trolleys in unit doses using the Kardex system was analysed retrospectively (n = 33,946 dispensings) and compared with the manual filling system (n = 5,284 dispensings); different filling procedures were used and compared. For each filling system and procedure, we obtained information for the number of dispensings and the time spent carrying this out. The speed of the process was calculated as the number of dispensings per hour. In addition, the drugs contained in the Kardex system were checked and the correspondence between both the actual and the theoretical content was assessed. RESULTS: The speed of the trolley filling process, expressed as the number of dispensings per hour, increased from 394 with the manual system (filling trolleys individually, n = 5,284 dispensings) to 417, 540 and 592 with the Kardex system when trolleys were filled individually (n = 16,530 dispensings), or grouped in two (n = 10,369 dispensings) or in three (n = 7,047 dispensings) respectively, at the start of the filling process. The check of the contents in the Kardex system detected 36% of trays with differences in quantity, one tray with out of date medication and three trays which should in theory have been empty and contained medication. CONCLUSIONS: The Kardex system optimises the speed of trolley filling with regard to the manual system, provided that the trolleys are grouped at the time of starting the preparation. The check of the cabinet contents shows that possibilities of error exist, related to its handling, which can be minimised by incorporating bar codes into the system s replenishment process.