Strategies for using pharmaceutical decision support systems and the potential contributions of AI in clinical pharmacy

Artificial intelligence (AI) is experiencing considerable growth in healthcare, including in clinical pharmacy. As medication review involves a large amount of health data from multiple sources, AI can be particularly useful in this field. Several commercial and in-house pharmaceutical decision support systems (PDSSs) have therefore been developed. The aim of this article is to describe the different strategies and organizations implemented by hospital pharmacists regarding their use and to analyze the other opportunities offered by AI for hospital pharmacy activities. Feedback was collected from five hospital pharmacists using different PDSSs for medication review in order to describe the organizations set up, the methods of use, and related uses in healthcare facilities. The opportunities for AI in this field were then analyzed. Feedback highlights the predominant use of PDSSs. It demonstrates that the deployment and use of PDSSs involves the mobilization of both time and pharmaceutical resources. Various advantages are described according to their functionalities and their ability to secure medication review: customization possibilities, real-time detection, and the ability to identify high-risk situations early on. However, these tools also have common limitations, such as data flow interoperability and alert fatigue. These limitations have helped identify professionals’ expectations regarding the improvement of such systems using AI-based technologies: the relevance of alerts, the integration of unstructured data, the optimization of pharmaceutical resources, facilitated rules development, risk prediction, clinical reasoning support, and continuity of care and/or community-hospital coordination. PDSSs therefore provide clear benefit to pharmacists, but improvements are still needed, while the integration of AI would open up new prospects and uses.