final_report.qmd

---
title: "Hands-On Example: Creating a Final Report with Quarto"
author: "Joshua J. Cook"
date: "`r Sys.Date()`"
format:
  html:
    toc: true
  pdf:
    toc: true
embed-resources: true
---

## Objective

The goal of this section is to use the complete CDISC-compliant SDTM and ADaM datasets to [**create a final report using Quarto**]{.underline}. This report will include [tables, listings, and figures (TLFs)]{.underline}, which are commonly required for regulatory submissions.

## Key Points for Presentation

-   **Overview**: Use the complete CDISC-compliant SDTM and ADaM datasets to create a final report using Quarto.
-   **Required Libraries**: `{tidyverse}` for data manipulation and figures, `{gt}` and `{gtsummary}` for listings and tables, and `{quarto}` for generating the final reproducible report.
-   **Input Data**: Imputed SDTM and ADaM Datasets.
-   **Data Understanding**: Understand how to create TLFs using imputed datasets to ensure compliance with regulatory standards.
-   **Report Creation Process**:
    -   Generate tables and listings using `{gt}` and `{gtsummary}`.
    -   Create figures using `{ggplot2}`.
    -   Compile the final report using `{quarto}`.

## Step-by-Step Report Creation with Quarto

### Step 1: Load Required Libraries

```{r}

if (!requireNamespace(c("tidyverse", "gt", "gtsummary", "quarto"), quietly = TRUE)) {
  install.packages(c("tidyverse", "gt", "gtsummary", "quarto"))
}

# Load necessary libraries
library(tidyverse)  # Tidyverse for data manipulation and figures
library(gt)         # GT for creating high-quality tables
library(gtsummary)  # Gtsummary for creating clinical trial summary tables
library(quarto)     # Quarto for generating the final report

# Loading the dataset in the final document
sdtm_dm_imputed <- readRDS("sdtm_dm_imputed.rds") # Loads our SDTM data 
adam_dm_imputed <- readRDS("adam_dm_imputed.rds") # Loads our ADaM data 
```

**Explanation**: We start by loading the `{tidyverse}` package for data manipulation and visualization, `{gt}` and `{gtsummary}` for creating tables and listings, and `{quarto}` for generating the final report. We also load in our previously generated SDTM and ADaM datasets.

### Step 2: Generate Summary Tables

```{r}
# Generate a summary table for demographic characteristics using gtsummary
summary_table <- sdtm_dm_imputed %>%
  # Select relevant columns for summarization: AGE, SEX, ETHNIC, and RACE
  select(AGE, SEX, ETHNIC, RACE) %>%
  
  # Create a summary table with grouping by SEX
  tbl_summary(
    by = SEX,  # Group data by SEX to summarize separately for each group
    statistic = list(all_continuous() ~ "{mean} ({sd})")  # Define how to summarize continuous variables (mean and standard deviation)
  ) %>%
  
  # Bold the labels for better visualization in the output table
  bold_labels()

# View the summary table
summary_table
```

**Explanation**: Here, we use `{gtsummary}` to create a summary table for demographic characteristics, grouped by `SEX`. The `tbl_summary()` function is used to summarize the `AGE`, `ETHNIC`, and `RACE` variables, with continuous variables reported as mean and standard deviation.

### Step 3: Generate High-Quality Tables for Listings

```{r}
# Generate a high-quality listing using gt
listing_table <- sdtm_dm_imputed %>%
  select(STUDYID, USUBJID, AGE, SEX, ETHNIC, RACE) %>%  # Select relevant columns for listing
  gt()  # Create a high-quality table using gt

# View the listing table
listing_table
```

**Explanation**: In this step, we use `{gt}` to create a listing of the demographic data. The `gt()` function helps generate publication-quality tables that can be included in the final report.

### Step 4: Create Figures Using ggplot2

```{r}
# Generate a histogram of AGE by SEX
age_histogram <- sdtm_dm_imputed %>%
  # Use the pipe operator to feed data into ggplot function for visualization
  ggplot(aes(x = AGE, fill = SEX)) +
  
  # Create a histogram with 'geom_histogram()'
  # 'aes()' defines the variables for the x-axis and fill color (by SEX)
  geom_histogram(binwidth = 5, alpha = 0.7, position = "dodge") +
  
  # Add descriptive labels for the title, x-axis, and y-axis of the plot
  labs(title = "Age Distribution by Sex", x = "Age", y = "Count") +
  
  # Apply 'theme_minimal()' for a cleaner, minimal visual style
  theme_minimal()

# View the histogram
age_histogram
```

**Explanation**: We use `{ggplot2}` from `{tidyverse}` to create a histogram of `AGE` by `SEX`. This figure helps visualize the distribution of age across different sexes in the dataset.

### Step 5: Compile the Final Report Using Quarto - hit RENDER

**Explanation**: Finally, we use `{quarto}` to compile the report. The `quarto_render()` function takes the input Quarto markdown file (`final_report.qmd`) and generates the output in HTML format, incorporating the SDTM and ADaM datasets.

## Important CDISC Concepts for Creating Regulatory-Compliant Reports

-   **Standardization**: The use of CDISC standards ensures that datasets are consistent across different clinical trials, which is essential for regulatory review.
-   **Traceability**: All derived variables, tables, and figures must be traceable to their source data, ensuring transparency.
-   **Metadata Documentation**: Proper documentation of variables and methods used in TLF generation is crucial for regulatory submissions.
-   **Regulatory Requirements**: TLFs are a key part of clinical trial reporting, and adhering to ADaM standards ensures that these outputs meet regulatory guidelines.

## Summary of Report Creation

-   **Tables and Listings**: Created summary tables and listings using `{gt}` and `{gtsummary}` to meet regulatory submission requirements.
-   **Figures**: Visualized key demographic data using `{ggplot2}`.
-   **Quarto Report**: Compiled all tables, listings, and figures into a final report using `{quarto}`.
-   **Compliance with CDISC Standards**: Ensured that all outputs were compliant with CDISC standards, providing traceability and proper documentation.

This hands-on section helps participants understand how to create a final clinical trial report that includes TLFs, emphasizing compliance with CDISC standards for regulatory submissions.