---
title: "NCC Introduction"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{NCC Introduction}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include=FALSE}
library(knitr)
opts_chunk$set(echo = FALSE,
               message = FALSE,
               error = FALSE,
               warning = FALSE,
               comment = "",
               fig.align = "center",
               out.width = "100%")

library(magick)
```



## Non-concurrent controls in platform trials

Platform trials are multi-arm multi-stage clinical trials that aim at evaluating the efficacy of several treatment arms within a single trial. Experimental treatments are permitted to enter and exit the trial at different times, as new treatments become available, and the total number of newly added treatments is not pre-specified. Moreover, treatment arms may be compared against a shared control group. 


For treatments that enter the trial later, the control group is divided into concurrent (CC) and non-concurrent controls (NCC), i.e., patients randomized before the given treatment arm entered the trial. 


```{r}
CC <- image_read("../man/figures/trial_scheme_CC.png")
NCC <- image_read("../man/figures/trial_scheme_NCC.png")
img <- c(CC, NCC)

image_animate(image_scale(img), delay = 150)
```


While using NCC data can offer several benefits, such as reduced sample size and increased statistical power, it may also introduce bias in effect estimators if time trends are present in the trial.