Generative AI and Large Language Models in Pharmacy Practice: A Scoping Review - Protocol

This project is a scoping review examining the application of generative artificial intelligence (GenAI) and large language models (LLMs) — including ChatGPT, GPT-4, DeepSeek, Claude, Gemini, Llama, Med-PaLM, BioBERT, ClinicalBERT, and GatorTron — in pharmacy practice settings. Background: The rapid emergence of LLMs has created significant opportunities and challenges across healthcare. Pharmacy practice, encompassing clinical pharmacy, hospital pharmacy, community pharmacy, and pharmaceutical care, represents a high-stakes domain where AI-assisted tools may improve medication safety, support clinical decision-making, and enhance patient counseling. However, the scope, quality, and limitations of existing evidence remain poorly characterized. Objectives: This scoping review aims to (1) map the existing literature on GenAI/LLM applications across pharmacy practice domains; (2) identify the primary use cases, performance outcomes, and limitations reported in the literature; and (3) identify gaps in current evidence to guide future research and implementation. Methods: We will follow the JBI methodology for scoping reviews and report findings according to the PRISMA extension for Scoping Reviews (PRISMA-ScR). A systematic search will be conducted across PubMed, Embase, Web of Science, and the Cochrane Library. Two independent reviewers will screen titles/abstracts and full texts. Data will be extracted using a standardized form and synthesized narratively. Expected Outcomes: This review will produce a comprehensive evidence map of GenAI/LLM use in pharmacy practice, highlight promising applications, identify methodological limitations in the current literature, and provide recommendations for future research and clinical implementation. Registration: This protocol is prospectively registered on OSF prior to commencement of data extraction.