Does it handle Bayesian adaptive methods?

Absolutely, it provides templates for Bayesian evidence synthesis, MAP priors, and robust prior construction using the RBesT package.

Which R packages does this skill support?

It supports major biostatistics packages including adaptr, rpact, MAMS, RBesT, basket, and gMCPLite for comprehensive trial design.

Is support for multiplicity included?

Yes, the skill includes graphical multiplicity adjustments using the gMCPLite package for complex testing hierarchies.

Can I simulate platform trials with this skill?

Yes, it includes detailed patterns for setting up and simulating platform trials, including rules for adding or dropping arms mid-trial.

Can this help with sample size re-estimation?

Yes, it includes implementation patterns for both blinded and unblinded sample size re-estimation based on interim data.

Advanced Adaptive Clinical Trial Designs

Name: Advanced Adaptive Clinical Trial Designs
Author: choxos

bychoxos

0•

Data Science & ML

Implements and simulates complex adaptive clinical trial designs using industry-standard R packages like adaptr, rpact, and RBesT.

This skill provides comprehensive support for biostatisticians and clinical researchers to design, simulate, and analyze advanced adaptive clinical trials in R. It covers a wide range of modern methodologies including platform, basket, and umbrella trials, multi-arm multi-stage (MAMS) designs, and Bayesian evidence synthesis. Users can leverage specialized templates for Thompson sampling, sample size re-estimation, and group sequential boundaries, enabling the development of more efficient and flexible trial protocols that can adapt based on accumulating interim data.

Key Features

01Simulation and setup of platform, basket, and umbrella trial designs

02Implementation of Response-Adaptive Randomization (RAR) using Thompson sampling

03Multi-Arm Multi-Stage (MAMS) trial planning for continuous and binary outcomes

040 GitHub stars

05Group sequential design analysis with O'Brien-Fleming boundaries and sample size re-estimation

06Bayesian meta-analytic predictive (MAP) priors and robust evidence synthesis

Use Cases

01Simulating basket trials to evaluate a single treatment across multiple patient subpopulations or biomarkers

02Calculating required sample sizes and stopping boundaries for trials with planned interim analyses

03Designing a platform trial that allows for adding or dropping treatment arms based on interim performance

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add choxos/biostatagent advanced-adaptive-trials

For use in Claude.ai and ChatGPT

Key Features

01Simulation and setup of platform, basket, and umbrella trial designs

02Implementation of Response-Adaptive Randomization (RAR) using Thompson sampling

03Multi-Arm Multi-Stage (MAMS) trial planning for continuous and binary outcomes

040 GitHub stars

05Group sequential design analysis with O'Brien-Fleming boundaries and sample size re-estimation

06Bayesian meta-analytic predictive (MAP) priors and robust evidence synthesis

Use Cases

01Simulating basket trials to evaluate a single treatment across multiple patient subpopulations or biomarkers

02Calculating required sample sizes and stopping boundaries for trials with planned interim analyses

03Designing a platform trial that allows for adding or dropping treatment arms based on interim performance