---
title: "Interactive Shiny Dashboard"
author: "Shubhram Pandey: Heorlytics Ltd"
date: "`r Sys.Date()`"
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{Interactive Shiny Dashboard}
  %\VignetteEngine{knitr::rmarkdown}
  %\VignetteEncoding{UTF-8}
---

```{r setup, include = FALSE}
knitr::opts_chunk$set(echo = FALSE)
```

## Overview

`htaBIM` ships with a fully interactive Shiny dashboard that lets you build,
explore, and export a budget impact model entirely through a point-and-click
interface — no R code required.

The app is built with **bslib** (Bootstrap 5) for a clean, modern layout and
works in any desktop browser.

---

## Launch locally

If you have the package installed, launch the app with one line:

```r
htaBIM::launch_shiny()
```

Required packages are listed in the *Suggests* field of `DESCRIPTION`; the
app will prompt you if anything is missing.

---

## Live demo

Deploy your own copy to shinyapps.io (see the Deployment section below) or
run the app locally with `htaBIM::launch_shiny()`.

---

## App structure

The dashboard is organised into six tabs:

| Tab | What it does |
|---|---|
| **Model Setup** | Enter all inputs: population funnel, up to 3 treatments, costs, payer, and uptake dynamics |
| **Results** | KPI cards (Year 1 / final year / cumulative), annual and cumulative line plots, interactive results table |
| **Market Shares** | Stacked bar chart comparing current vs. new-drug scenario across years |
| **Sensitivity** | One-way DSA with tornado diagram + Probabilistic Sensitivity Analysis (Monte Carlo) |
| **Scenarios** | Side-by-side scenario comparison table and per-patient cost breakdown chart |
| **Report** | Plain-text and HTML report download |

---

## Model Setup tab

### Population funnel

| Input | Description |
|---|---|
| Indication / disease | Free-text label for the condition |
| Reference country | Selects built-in population data (GB, US, CA, DE, FR, IT, AU, JP) or *custom* entry |
| Prevalence | Point prevalence as a proportion of the reference population |
| Diagnosed rate | Proportion of prevalent patients with a diagnosis |
| Treated rate | Proportion of diagnosed patients receiving any treatment |
| Eligible rate | Proportion of treated patients eligible for the new drug |
| Growth rate | Annual growth in the eligible population |
| Projection horizon | Number of years (1–10) |

### Treatments

Up to **three treatments** are supported:

- **New drug** — name, target market share at end of horizon (%), and uptake dynamics
- **Comparator 1** — always required; represents the main standard of care
- **Comparator 2** — optional second comparator (tick the checkbox to enable)

Four uptake dynamics are available: *linear ramp*, *logistic (S-curve)*,
*step change*, and *constant*.

**Scenario variants** (conservative / optimistic peak shares) are defined in
this tab and propagate through all analyses.

### Costs & payer

| Input | Description |
|---|---|
| Drug costs | Annual list price per treatment |
| Administration cost | Annual infusion / injection cost for new drug |
| Monitoring cost | Annual monitoring cost for new drug |
| AE cost | Expected annual adverse event cost for new drug |
| Rebate | Confidential price discount on new drug (%) |
| Payer perspective | Generic healthcare system, NHS England, CADTH, or US commercial |
| Currency | GBP, USD, EUR, CAD, AUD, JPY |
| Discount rate | Applied to Year 2+ costs |

---

## Sensitivity tab

### Deterministic sensitivity analysis (DSA)

Set low/high ranges for:

- Prevalence
- Diagnosed rate
- Eligible rate
- New drug cost multiplier

Then click **Run DSA**. A tornado diagram and interactive results table are
shown for the selected year.

### Probabilistic sensitivity analysis (PSA)

Set standard errors (for epidemiology rates, sampled from Beta distributions)
and a coefficient of variation (for costs, sampled from LogNormal
distributions), choose the number of simulations, and click **Run PSA**.

A histogram of simulated budget impacts is shown with the base-case value and
95 % credible interval marked.

---

## Deployment to shinyapps.io

To host your own public copy:

1. Install `rsconnect`:
   ```r
   install.packages("rsconnect")
   ```

2. Connect your shinyapps.io account:
   ```r
   rsconnect::setAccountInfo(
     name   = "<your-account>",
     token  = "<token>",
     secret = "<secret>"
   )
   ```

3. Deploy the bundled app:
   ```r
   rsconnect::deployApp(
     appDir  = system.file("shiny/htaBIM_app", package = "htaBIM"),
     appName = "htaBIM"
   )
   ```

4. Copy the resulting URL and add it to your vignette or dossier.

---

## Linking from pkgdown

Once the pkgdown site is deployed to GitHub Pages (via the supplied GitHub
Actions workflow), this vignette renders as an article at:

```
https://<org>.github.io/htaBIM/articles/shiny-app.html
```

You can embed the shinyapps.io link in any article, the `README`, or the
`_pkgdown.yml` navbar.