Risk analysis of the non-sterile compounding activity and the impact of computerization in the preparation circuit

Introduction
Currently, the management of our processes contributing to the manufacturing of non-sterile preparations depends on several tools: paper (fax), emails, prescription software, Excel^® and Word^® files. These tools are not completely secure and can be sources of errors when carrying out a pharmaceutical preparation. A typical FMEA risk analysis was carried out. The contribution of a single software in managing these risks was also evaluated.

Method
The various risks are identified based on risk analyzes previously carried out in the service, requests for non-compliance and observation of practices. The processes analyzed include: demand analysis, inventory management, preparation, dispensing as well as hygiene, safety and equipment. Gross criticality is defined as the product of the frequency of occurrence of the risk, its severity and its detectability. These criteria were assessed by the entire team (n=13, preparers, pharmacists and interns) via a questionnaire. The gross criticality is weighted by a factor expressing the means of controlling the risk, this results in a net criticality. Criticalities are classified into 3 levels: C1 “acceptable as is”, C2 “to be monitored” and C3 “to be treated as a priority”. To evaluate the impact of a single software, we assumed functionalities corresponding to our current practices to estimate the reduction in risk criticality.

Results
70 risks were identified. Concerning their gross criticality, 20% are classified C1, 54% C2 and 26% C3. Thanks to the means of control, the net criticality is assessed: 53% of the risks are then classified in C1, 36% in C2 and 11% in C3. Evaluation of the unique software showed that it could further reduce risk criticality through process automation, reduction of manual transcriptions and improved traceability. With the implementation of this software, the risks would be reclassified as follows: 71% in C1, 20% in C2 and 9% in C3.

Conclusion
Although our system uses non-specific computerized tools, it is not sufficiently secure, with transcriptions being sources of errors. The use of a single software could significantly reduce these risks by automating processes and improving traceability. This would contribute to securing the drug circuit and the dematerialization of our tools, in accordance with Good Preparation Practices.

See the poster