PIVAS quality and safety management based on 6S management informatization construction

The traditional pharmacy intravenous admixture services (PIVAS) faces challenges in the efficiency and accuracy of medical order processing. This study proposes and evaluates a novel PIVAS quality and safety management system integrating 6S management with a Graphical Evaluation Review Technique (GERT) and Interval Rough Number (IRN) model. The GERT-IRN model demonstrated low computational load (14.6 MFLOPs), high accuracy (98.65%), and high recall (96.0%) in simulation and clinical tests. More importantly, a two-month clinical trial at a certain hospital (n=63,182 orders) confirmed its real-world effectiveness. The system increased the interception rate of unreasonable medical orders to 95.9% (95% CI 95.4–96.4%), successfully blocking 9 high-risk contraindications missed by manual review. Furthermore, it dramatically reduced the cytotoxic drug dispensing timeout rate from 18.3 to 4.1% (p < 0.001), a 77.6% improvement. Overall, the optimized PIVAS system enhances the efficiency, safety, and effectiveness of patient medication, providing strong technical support for the intelligent improvement of medical quality.