RWEhub 🧬📊

A curated tutorial hub for Real-World Evidence (RWE), built for the CABS RWE Internship Program.

RWEhub is a teaching companion that summarizes and organizes the open-source projects tagged under GitHub's rwe topic. Instead of a flat list of stars, it groups the ecosystem into a guided learning path so an intern with a stats/programming background can go from "What is real-world evidence?" to running a reproducible RWE study of their own.

Who this is for: CABS RWE interns, new analysts, and anyone learning pharmacoepidemiology, OMOP/CDM data engineering, and causal inference on real-world healthcare data.

---

What is Real-World Evidence (RWE)?

Real-World Data (RWD) is health data collected outside of randomized controlled trials — from electronic health records (EHR), insurance claims, disease registries, patient-reported outcomes, and wearables.

Real-World Evidence (RWE) is the clinical evidence about the use, benefits, and risks of a medical product derived from analyzing RWD. Regulators (FDA, EMA) increasingly accept well-designed RWE to support approvals, label expansions, and post-market safety.

A modern RWE study almost always involves four skills, which map to the four tracks in this hub:

Skill	Question it answers	Track
Data engineering	How do I turn messy EHR/claims data into an analyzable format?	Track 1: Data & OMOP CDM
Reproducibility	How do I make my study auditable and repeatable?	Track 2: Reproducibility & Study Lifecycle
Causal inference	How do I estimate a treatment effect without a randomized trial?	Track 3: Methods & Causal Inference
Application	What does a real end-to-end study look like?	Track 4: Applied Case Studies

---

🗺️ The 4-Week Intern Learning Path

Each week pairs a concept with a hands-on repository to clone and run.

Week 1 — Foundations & Data

Understand RWD sources and the OMOP Common Data Model (CDM), the standard that lets studies be portable across databases.

• Read: What is RWE · Glossary
• Do: Stand up a lakehouse and explore an OMOP schema — databricks-industry-solutions/omop-cdm, eyedress02/modern-lakehouse

Week 2 — Reproducibility

Learn version control, environment locking (renv), and audit reporting so your work survives peer and regulatory scrutiny.

• Do: Work through janickweberpals/icpe-git-2024 (git for the RWE lifecycle)

Week 3 — Methods

Propensity score matching, IPTW, Cox proportional hazards, Kaplan–Meier survival, and causal inference.

• Do: Run the end-to-end pipeline in repro-stats/reproducr-rwe; study al8xi8/EPID708_UMich2025

Week 4 — Apply

Reproduce a domain study and start your own mini-project.

• Do: Explore htlin222/roche-vabysmo-rwe-workshop, Krashnika-Deivakumar/aml-flt3-market-sizing, natsousa/rwbiomarker
• Then: Pick an Internship Project

Full curriculum with checkpoints → docs/01-learning-path.md

---

📚 Curated Repository Catalog

The 15 repositories from the rwe topic, grouped by what they teach. Full annotations in docs/02-curated-repos.md.

Track 1 — Data & OMOP CDM

Turning raw health data into standardized, analyzable tables.

Repo	Lang	What you'll learn
databricks-industry-solutions/omop-cdm	Python	Building an OMOP CDM on a modern data lakehouse ⭐ start here
eyedress02/modern-lakehouse	Shell	Lakehouse stack with MinIO, MariaDB, Dremio
ericg1212/healthcare-claims-pipeline	Python	HL7 FHIR → OMOP CDM, claims classification, cohort building
aminyakubu/explorys	R	Pulling cohort demographics from the IBM Explorys database

Track 2 — Reproducibility & Study Lifecycle

Making studies transparent, auditable, and repeatable.

Repo	Lang	What you'll learn
janickweberpals/icpe-git-2024	HTML	Git + transparency across the RWE study lifecycle ⭐ start here
repro-stats/reproducr-rwe	R	Full reproducible pipeline: PS matching, IPTW, Cox, KM, renv, audit report
janickweberpals/agentic-code-review-ispe-2026	R	Using AI agents to review analytics code

Track 3 — Methods & Causal Inference

The statistics that make RWE credible.

Repo	Lang	What you'll learn
al8xi8/EPID708_UMich2025	R	Machine learning + causal inference for epidemiology (course)
htlin222/roche-vabysmo-rwe-workshop	R	Interactive textbook: clinical RWE analysis with R + AI

Track 4 — Applied Case Studies & Portfolios

End-to-end and domain-specific examples to model your own work on.

Repo	Lang	What you'll learn
Krashnika-Deivakumar/aml-flt3-market-sizing	R	RWD simulation for AML market sizing + survival analysis
natsousa/rwbiomarker	R	RWE biomarker analysis in colon cancer
kousha1234/Research-portfolio	R	R Markdown research portfolio on synthetic data
kshptl/kshptl.github.io	JS	Data-scientist portfolio site (presenting your work)
QuartzSoftwareLLC/quarry	HTML	Data insight & visualization dashboard

ℹ️ baali-who/rwaise also carries the rwe tag but is a blockchain ("Real-World Asset") project — not real-world evidence. Included for completeness; skip it for RWE learning.

---

🚀 Quick Start for Interns

# 1. Clone this hub
git clone https://github.com/cabsweb/RWEhub.git
cd RWEhub

# 2. Read the orientation docs
open docs/00-what-is-rwe.md
open docs/01-learning-path.md

# 3. Follow Week 1 — clone your first data repo
git clone https://github.com/databricks-industry-solutions/omop-cdm.git

🗂️ Repository Contents

RWEhub/
├── README.md                     # You are here
├── docs/
│   ├── 00-what-is-rwe.md          # RWE/RWD concepts & data sources
│   ├── 01-learning-path.md        # Detailed 4-week curriculum + checkpoints
│   ├── 02-curated-repos.md        # Full annotations for all 15 repos
│   ├── 03-glossary.md             # OMOP, IPTW, Cox PH, FHIR, ... defined
│   └── 04-internship-projects.md  # Starter project ideas + rubric
├── CONTRIBUTING.md                # How interns add repos & fix docs
└── LICENSE

🤝 Contributing

Interns are encouraged to keep this hub current. Found a new RWE repo, or a better explanation? See CONTRIBUTING.md.

---

Maintained by the CABS RWE Internship Program · Source topic: https://github.com/topics/rwe