DD-006 Comparative heuristic evaluation between two versions of a computerised physician order entry system

BackgroundComputerisation of drug prescription is an essential part of securing the management of patient care. However, errors related to computer misuse can be a source of iatrogenic injuries. These errors can be partly attributed to usability problems of computerised physician order entry (CPOE).PurposeIn 2014, our hospital started computerised prescription on the CPOE ORBIS NICE developed by AGFA Healthcare. Given the difficulties encountered during ORBIS NICE implementation, a new ergonomically redesigned version called ORBIS Mobile Edition (ME) was proposed in late 2015. The aim of this study was to compare the usability of drug prescription in these two versions.Material and methodsTwo pharmacists conducted a heuristic evaluation based on drug prescription in an orthopaedic surgery ward. They analysed all the screens appearing to prescribers during the prescribing process, from the connexion to the software to the prescription’s signing. Each usability problem (UP) observed was analysed based on the stage of the process according to ergonomic criteria defined by Bastien and Scapin.1 Then, a severity score was assigned to each UP from 1 (non-essential resolution) to 4 (catastrophic, imperative to fix).Results97 UPs were detected on ORBIS NICE with 70 (72%) strictly on the prescription part. The total average severity of these UPs was 3.0, considered a major severity. Catastrophic severity UPs were especially related to ‘drug search’. During the drug’s selection, the system sometimes automatically moved without the user’s control which increased the workload and the probability of making errors. Only 44 UPs were identified on ORBIS ME, including complete resolution of UP on stage ‘drug search’. However, 29 new UPs of average severity of 2.8 were observed on this version and the total average severity was the same between the two CPOE.ConclusionThe heuristic evaluation has highlighted a relative improvement in the ergonomics of ORBIS ME. However, several UPs with major to catastrophic severity have not been corrected and new UPs appeared in this version. Other user tests are intended to assess the satisfaction of prescribers with this new version.References and/or acknowledgements1. Bastien JMC, Scapin DL. A validation of ergonomic criteria for the evaluation of human-computer interfaces. Int J Hum Comput Interact1992;4:183–96.No conflict of interest