---
title: "summaryByVisit"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{summaryByVisit}
  %\VignetteEncoding{UTF-8}
  %\VignetteEngine{knitr::rmarkdown}
editor_options: 
  markdown: 
    wrap: 72
---

```{r, include = FALSE}
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  echo = TRUE,
  eval = TRUE,
  warning=FALSE,
  fig.height = 6,
  fig.width = 9,
  fig.align='center'
)
```

```{r color-function, echo = FALSE}
colorize <- function(text, color) {
  if (knitr::is_latex_output()) {
    sprintf("\\textcolor{%s}{%s}", color, text)
  } else if (knitr::is_html_output()) {
    sprintf("<span style='color: %s;'>%s</span>", color, text)
  } else text
}

```

```{r setup, echo = FALSE, message = FALSE, warning = FALSE}
library(dplyr)
library(tidyverse)
library(gtsummary)
library(summarySCI)
library(flextable)
```

The function `summaryByVisit()` produces a table with descriptive statistics for continues variables at different time points (visits). It is largely based on the function `gtsummary::tbl_summary()` and `gtsummary::tbl_strata_nested_stack`. The changes as
compared to these functions are:

-   A hierarchical summary table is created by variable (lab-value), visit and visitgroup.

## Setup and data

To demonstrate the various functionalities of the function, we will create a small
dataset. We have three hypotetical lab-values: LDH, ANC and Lymphocytes. Each lab-value has been measured at 10 time points (visits), which can further be divided into three visit groups (baseline, treatment and follow-up). In addition, we have a grouping variable called 'arm'.

```{r, message = FALSE}
data<-NULL
visit <- c(paste0(rep("Visit ", 10), rbind(c(1:10))),
           paste0(rep("Visit ", 10), rbind(c(1:10))),
           paste0(rep("Visit ", 10), rbind(c(1:10))))
data <- as.data.frame(cbind( visit, rnorm(30)))
data<-as.data.frame(rbind(data, data, data, data, data))
data$visitgroup<- ifelse(data$visit %in% c("Visit 1", "Visit 2"), "Baseline", ifelse(data$visit %in% c("Visit 3", "Visit 4"), "Treatment", "Follow-up"))
data$visitgroup<-factor(data$visitgroup, levels = c("Baseline", "Treatment", "Follow-up"))
data$LDH<-rnorm(150)
data$Lymphocytes<-rnorm(150)
data$ANC<-rnorm(150)
data$LDH[3]<-NA
data$arm<- c(rep("Arm A", 70), rep("Arm B", 80))
```

## Basic table

Now, we use `summarySCI::summaryLevels` to create a summary table for median and range of each lab-value by visit.

```{r}
summaryByVisit(data,
         vars = c("LDH", "Lymphocytes", "ANC"),
         visit = "visit", 
         add_n = TRUE)
```
## Add visit group
We can the visit group to group visits accordingly. Visitgroup needs to be an ordered factor.

```{r}
summaryByVisit(data,
         vars = c("LDH", "Lymphocytes", "ANC"),
         visitgroup = "visitgroup",
         visit = "visit")
```

## By group 
We can stratify the table by groups via the `group` argument. The overall column can still be shown if desired, using the `overall = TRUE` argument. A maximum of 3 groups are supported.

```{r}
summaryByVisit(data,
         vars = c("LDH", "Lymphocytes", "ANC"),
         group = "arm",
         visitgroup = "visitgroup",
         visit = "visit")
```

```{r}
summaryByVisit(data,
         vars = c("LDH", "Lymphocytes", "ANC"),
         group = "arm",
         overall = TRUE,
         visitgroup = "visitgroup",
         visit = "visit")
```

## Add N

Sample size can be shown for each column, if the option `add_n` is set to `TRUE`.

```{r}
summaryByVisit(data,
         vars = c("LDH", "Lymphocytes", "ANC"),
         group = "arm",
         overall = TRUE,
         visitgroup = "visitgroup",
         visit = "visit",
         add_n = TRUE)
```

# Further steps:

-   improve costumization