Talks & Presentations
πΉ ShinyConf 2025
Title: Reviewing Clinical Data Efficiently with Shiny
π Date: 2025-04-12
π Summary:
This presentation introduces a Shiny-based application designed to improve the efficiency of clinical data review. Traditional EDC systems often limit reviewers to viewing data one form and one patient at a time, making it difficult to cross-reference information across forms such as Adverse Events (AE), Exposure (EX), and Concomitant Medication (CM). This tool addresses that challenge by providing a user-friendly, click-driven interface that allows reviewers to select patients, filter forms and variables, and instantly visualize clinical timelines and data listings. The application maintains the original data structure, requires minimal setup by programmers, and is accessible to non-programming users. Key benefits include simultaneous multi-form data review, integrated visualizations and listings, and Excel export functionality. This tool aims to bridge communication gaps between reviewers and programmers while enhancing the speed and clarity of clinical review workflows..
πΉ R/Pharma 2024
Title: Using Shiny to Clearly Present Clinical Results with CDISC-Compliant Dataset
π Date: 2024-10-31
π Summary:
This presentation explores how R and Shiny can enhance the review and visualization of clinical trial data. Traditional workflows often involve repeated back-and-forth verification between datasets such as SDTM, ADaM, and EDC, which is time-consuming. By leveraging Rβs Shiny framework, we can streamline data filtering and visualization, enabling faster and more intuitive review processes for both statisticians and medical teams. The talk highlights three key Shiny applications: tumor response visualization, patient milestone tracking, and SDTM domain review. These tools support both population-level summaries and individual-level insights. Emphasis is also placed on the importance of using CDISC-compliant data formats to standardize and simplify data handling. Finally, the integration of Shiny with Quarto is introduced as a future direction to make clinical data more accessible to non-programmers, improving data transparency and efficiency in reporting.
πΉ Pharmasug 2018
Title: Using Shiny to Clearly Present Clinical Results with CDISC-Compliant Dataset
π Date: 2018-08-31
π Summary:
This paper discusses how to enhance programming quality and efficiency in clinical trials under tight deadlines, especially within CROs. It emphasizes the importance of βfirst-time quality,β defined as the quality of deliverables before QC review. The author argues that poor initial quality leads to higher correction costs and delays, despite common practices like SOPs, validated macros, and training. A structured QC plan is essential but resource-intensive. The paper highlights factors affecting quality, including unfamiliarity with study design, insufficient task understanding, and client complexity. To address this, the author proposes a βFirst-Time Quality Scaleβ that evaluates programmers based on QC comments, client feedback, and satisfaction. High-scoring individuals can be assigned to time-sensitive projects, maximizing both quality and speed. The paper concludes that focusing on first-time quality can reduce repetitive work, save resources, and improve overall outcomes, aligning with the increasing demand for rapid and accurate clinical data processing.