Category: Professional Posters
Purpose: Optimizing automated dispensing cabinets (ADCs) and minimizing labor resources remains a challenge. This study aims to compare, describe, and benchmark key performance indicators (KPIs) after utilization of inventory optimization predictive analytics and to subsequently estimate the potential financial impact at acute care health system pharmacies.
Methods: A retrospective database analysis to evaluate key performance indicators (KPIs) pre and post implementation of enterprise level inventory management with predictive analytics was performed. We focused on the percentage of stock outs KPI in the patient care and procedure areas.
All dispensing records from the ADCs available for research were aggregated using de-identified data between 2016 and 2019. Researchers were blinded to any information related to hospital name, location, number of beds, and any transactional information. KPIs were aggregated and analyzed from the queried data. Descriptive statistics and trends on these performance metrics were generated using Microsoft Excel.
Results: This retrospective data analysis included over 30 acute care health system pharmacies pre- and post-implementation of inventory optimization predictive analytics. Stock outs identify the areas with potential for improvement in inventory management, therefore improving the medication availability to nursing.
The median stockout rate pre-implementation was 1.54% (0.34% standard error) versus median stockout rate post-implementation of 0.74% (0.26% standard error). The difference in pre- and post-implementation stockout rates was statistically significant at 0.8% (p<0.05).
Carrying too little inventory can create supply exhaustion (incidents of depleted stock at specific locations) that can result in a missing dose and delay in medication administration.
Conclusion: Inventory optimization is key to maximizing the benefits of ADC technology. This retrospective database analysis demonstrated that inventory optimization predictive analytics, which can recommend PAR level modifications and prioritize medication management, can potentially be used to drive reductions in stock outs and expired medication waste as well as optimize inventory storage. With continued rises in medication costs, medication inventory management is increasingly important. Big data can be leveraged to (1) help individual facilities optimize their technology (2) benchmark facilities across health-systems.