Training Contents for Shiny-based Teal App

### Feature description

**1. Concept of Teal (20 seconds)**
R shiny-based framework to generate data insights and present it in a layout similar like traditional clinical outputs but in an interactive web interface.
**2. Business Examples** ( priority): Recommend to 1) introduce the purpose and importance of outputs in each category and show a few examples: Efficacy > Safety > Patient Profile > Explortary (one min)

**Efficacy (1-2 mins): Interpretation of the drug efficacy (ready in teal. gallery)**

- The classic statistical methodologies and rationale for Oncology.
- KM and HR: to see the active drug and placebo over time (Show Exp 1)
- Forrest Plot: to see if some subgroup has drug efficacy in a different direction (Show Exp 2)

- Traditional therapeutical area:
- Response rate (Show Exp 3)

**- Business case for Teal:** want to see what the results look like for subgroups or different combinations of subgroups which may indicate the new signal ( Show Exp 1 but filter on subgroup)

**Safety(30 secs):** there are examples in teal. Gallery and since these parts is traditional and not time-consuming even generated from traditional SAS, so would recommend showing that teal can do that and giving an overview of a few tables

**Patient Profile ( 2 mins):** This part I feel has the most potential value since patient profile questions are always asked by scientists or external physicians when we have safety review meetings.

- Go through each part of the patient profile ( 1 min)
- Give an example of the use case of one or two particular examples ( 1 min): for example, during a JMC safety review meeting, if the reviewers want to see a particular patient's narratives like if he or she has grade 4 /5 AEs, what his her MH, and other AEs, what his/her lab tests over time, this will help science make a fair judgment if the grade 4 or 5 AEs are due to drug safety or due to the condition of patients him/herself

**Exploratory:** I saw some good plots in this part, my experience is that no scientist has ever asked similar analysis in late once so not sure if this is more specific rather than general outputs. Maybe give an overview in 30 secs, and explain what scientific questions this will help answer.

**3. Line by line Coding with the explanation of the purpose behind the coding**

- Show the line-by-line coding to a few examples above (recommend, KM, Forrest Plot, Response tabel, Patient Profile of lab and AE, one plot from exploratory)

- @telepath37