Risk analysis of the chemotherapy production process: what changes after the deployment of the Drugcam® control tool?
CHD Vendée, Bd Stéphane Moreau, 85925 La Roche sur Yon
Context and objectives
A risk analysis regarding the preparation of intravenous chemotherapy was conducted in 2015 and led to the identification of 33 Failure Modes (FM) among which 5 were qualified as unacceptable. Drugcam®, a video system designed to control the chemotherapy preparations, was then identified as the main corrective action to 4 of the 5 unacceptable FM. The aim of this study is to assess the impact of the deployment of Drugcam® in 2018.
A Failure Mode and Effects Analysis (FMEA) was conducted before and after the deployment of Drugcam®. This method consists of a multidisciplinary work group identifying the potential failures of the process of chemotherapy preparation, as well as their causes and consequences. 33 FM were identified among 6 stages of the process. The criticality index was calculated for each FM based on three criterias: gravity, frequency and detectability. The residual criticality of the FM was calculated by weighting the initial criticality by existing risk management means. Each FM is classified according to their residual criticality in one of the three following risk categories: acceptable, tolerable under control and unacceptable. The criticality difference between the two FMEA enabled us to evaluate the risk evolution of each FM.
The overall criticality decreases by 20%. Of the 33 FM identified, the criticality after the deployment of Drugcam® decreases for 13 FM (39%), remains unchanged for 9 FM (27%) and increases for 11 FM (33%). Between the two FMEA, the unacceptable FM decrease from 5 (15%) to 1 (3%) and relate to the risk of medication error and medical device error or their amount. The number of tolerable FM under control is stable (27%) and acceptable FM increases from 17 (51%) to 21 (63%).
The deployment of Drugcam® has a major impact on the risk management of chemotherapy preparation. Drugcam® enables to reduce the overall risk criticality and 3 of the 4 unacceptable FM to become acceptable. The remaining unacceptable FM is explained by a new profile of errors relying on the user’s confidence in the tool that can lead to control failures regarding certain key elements (operating mode, lot number,...). This tool may also be responsible for increasing the criticality of some FM such as delayed administration related to the preparation’s release (failure to prioritize and increased release time). This risk analysis has helped to develop improvements of the system with the publisher and to improve the users’ training. Drugcam®, like any automated system, induces specific risks that need to be identified and controled. Nevertheless, it allows a significant improvement in the process by detecting errors and allowing their analysis, which is essential to train its users.