An “R-Shiny” Interface Designed As a One-Stop Solution for All Kinds of Survival Analysis According to NICE Technical Support Documents 14 and 21

OBJECTIVES: Survival modeling is essential for Health Technology Assessment (HTA) bodies to evaluate treatments impacting survival and health-related quality of life in health economic evaluations. Numerous software tools and open-source R packages are available for conducting survival analysis in HTA submissions, including extrapolation beyond follow-up time points. This project aims to develop an "R-shiny" software interface to standardize survival analysis, providing a comprehensive solution for both conventional and advanced survival analysis methodologies.

METHODS: The user-friendly web application, developed with the R Shiny framework, is deployed using Docker containers on Amazon Web Services (AWS) to ensure accessibility and scalability. User security is maintained with SSL certificates and Auth0 authentication. The application accepts individual patient-level (IPD) time-to-event data or digitized data (pseudo-IPD using Guyot algorithm will be generated by application) as input. Users can select the type of analysis to perform via dropdown menus, which include fitting standard parametric models (Exponential, Weibull, Log-normal, Log-logistic, Gompertz, Gamma, and Generalized Gamma), spline-based models (odds, hazard, or probit scale) with up to three knots, cure models, piece-wise models, joint-fitted models, and parametric mixture models.

RESULTS: The application creates an interactive dashboard for all fitted models, displaying goodness-of-fit statistics, Kaplan-Meier plots versus fitted plots, and a milestone estimates table with landmark probabilities. It also summarizes time-to-event data, including proportional hazard assumption testing, Schoenfeld residual plots, and quantile-quantile plots. Additionally, the application generates estimates in Excel format for economic models and produces dynamically updated Word reports for documentation purposes.

CONCLUSIONS: The developed "R-shiny" interface (Survlytics) standardizes survival analysis, offering a comprehensive, user-friendly solution for both conventional and advanced methodologies. It enhances accessibility, ensures data security, and provides interactive dashboards with reports, making it a valuable tool for HTA submissions. Additionally, this tool can be enhanced with Generative AI for a QA chatbot, offering deeper insights into the data.