Scope of this User Guide

This document covers the design options that are common across the four FACTS Core Design Engines: Continuous, Dichotomous, Time-to-Event, and Multiple Endpoint. Some design elements are shared across design engines, in which case there is only a single description of them. Others differ based on the endpoint used, in which case separate pages have been created to describe each.

The screenshots provided are specific to a particular installation and may not reflect the exact layout Screenshots from earlier versions of FACTS 6 are still used only when the tabs they show are unchanged in FACTS 7.1. of the information seen by any particular user. They were taken from FACTS V7 & V6 installed on Windows 10 or 11. If FACTS is installed on different versions of Windows, or with different Windows ‘themes’ there will be some differences in appearance introduced by Windows. The contents of each tab, however, will generally be consistent with the screenshots in this document.

FACTS Version

This is the version of the user guide for inclusion with the FACTS 7.1 release.

Citing FACTS

Please cite FACTS wherever applicable using this citation.

FACTS Core Overview

FACTS Core is generally used to simulate trials where one or more active arms are compared to a common control arm. Examples of trials that can be simulated in FACTS Core engines are: Group sequential designs, goldilocks designs, dose- finding phase 2 designs, fixed phase 3 designs, arm dropping designs, response adaptive allocation designs, and more.

FACTS Core can be frequentist or Bayesian, and as adaptive as needed or fixed, and not adaptive at all.

FACTS provides numerous options for the statistical analysis, some of them straight forward, some of them quite novel, though all have been used in actual trials:

The endpoint can be continuous, dichotomous, or time to event. Multiple continuous/dichotomous endpoints can be simulated in 1 trial.
As well as a control arm, the study arms can be compared with an active comparator.
When simulations are executed, they can be executed for a wide range of scenarios. A scenario can be composed of a combination of one of each type of profile – dose response, longitudinal, accrual, and dropout.
With either a dichotomous or continuous endpoint, longitudinal models can used to impute the patient’s likely final outcome from early interim measures. This can be used when final endpoint data is missing due to subject drop-out or at interims for subjects who’ve not reached their final endpoint yet.
Estimation of the response on the control arm can be augmented using a hierarchical model to borrow from data from previous studies (This is known as Bayesian Augmented Control, BAC).
Interim analyses can be specified at fixed intervals by time, the number of subjects recruited, or the number of events observed.
At interim analyses, options include:
- Choosing to stop the whole study for success or futility.
- Dropping treatment arms
- Adapting the randomization proportions to favor allocating to the doses that are most likely to be the desired target – which can be the study arm with the maximum response, the EDx, or minimum efficacious dose.

The main tabs in FACTS:

The Study Tab: The Study Tab is used for entering the main characteristics of the study and specifying the treatment arms to be tested in the study. This is where the user specifies the ‘given’ requirements, or constraints, of the trial to be designed.
Virtual Subject Response: The VSR tabs are for specifying the simulation of subject responses. This is either by specifying assumed efficacy scenarios for each treatment arm, or by supplying external data files of simulated subject responses.
Execution: The Execution tab is for specifying operational properties of the trial’s execution to simulate. Specifically, the rate of subject accrual and the probability of subjects dropping out.
Quantities of Interest: The Quantities of Interest tab is for specifying the quantities to be calculated during analyses for output in the results, and possible use in adaptive allocation, early stopping and final evaluation decisions.
Design: The Design tabs are for specifying the statistical analysis and adaptive design rules. These are the design choices available to the trial biostatistician and include how to model the final responses, allocation rules, missing or unobserved data handling, what decisions to make at interims, and how to determine if the overall trial has been successful.
Simulation: The simulation tab is for the user to set up and run simulations. It also allows for viewing of the simulation results either as output simulation files or pre-made graphics that provide many summaries of the simulation outcomes.
Analysis: On the Analysis tab, the user can load an example data set and view the results of the FACTS analysis of that data using the current specified design. This tab is useful for implementation of a design that was simulated in FACTS.

Many of the above tabs also has sub-tabs that help with configuring the specifics of the design simulations. These sub-tabs are described in detail in the tab- specific portions of the user guide.