Real World Evidence MCP Server

RWE Life Sciences MCP Server

A local Model Context Protocol (MCP) server that exposes a single Real-World Evidence (RWE) tool to Claude. The tool returns aggregated, de-identified patient cohort statistics across four dimensions:

Dimension	Values
Age group	0-17, 18-34, 35-49, 50-64, 65-74, 75+
Gender	Male, Female, Unknown/Other
US Geography	4 Census regions + all 50 states + DC
Longitudinal	2018 – 2024 (year-over-year trend)

Disclaimer – All data is synthetically generated for demonstration purposes only. Do not use for clinical or regulatory decisions.

---

Quick start

1. Install dependencies

cd rwe_mcp_server pip install -r requirements.txt

2. Run the server

python server.py

The server starts on Streamable HTTP at:

• MCP endpoint: http://127.0.0.1:9090/mcp
• Health endpoint: http://127.0.0.1:9090/health

You can override the bind address, port, or transport:

MCP_HOST=0.0.0.0 MCP_PORT=9090 python server.py MCP_TRANSPORT=stdio python server.py

---

Connect to Claude Desktop

Open (or create) claude_desktop_config.json:

OS	Path
macOS	~/Library/Application Support/Claude/claude_desktop_config.json
Windows	%APPDATA%\Claude\claude_desktop_config.json

Add the server entry below (replace /absolute/path/to/ with your actual path):

{ "mcpServers": { "rwe-life-sciences": { "command": "python", "args": ["/absolute/path/to/rwe_mcp_server/server.py"], "env": { "MCP_TRANSPORT": "stdio" } } } }

If you use a virtual-environment interpreter:

{ "mcpServers": { "rwe-life-sciences": { "command": "/absolute/path/to/.venv/bin/python", "args": ["/absolute/path/to/rwe_mcp_server/server.py"], "env": { "MCP_TRANSPORT": "stdio" } } } }

Restart Claude Desktop. The 🔧 tool icon should show rwe-life-sciences connected.

---

Tool reference

query_rwe_patient_cohort

Parameters

Parameter	Type	Required	Description
condition	string	✅	Clinical condition (see list below)
dimensions	string[]	❌	Subset of ["age","gender","geography","longitudinal"]. Default: all four.
year_start	int	❌	Start of observation window (2018–2024). Default 2018.
year_end	int	❌	End of observation window (2018–2024). Default 2024.
region	string	❌	US Census region filter: Northeast, South, Midwest, West.

Supported conditions

• Type 2 Diabetes
• Hypertension
• Heart Failure
• COPD
• Asthma
• Atrial Fibrillation
• Major Depressive Disorder
• Rheumatoid Arthritis
• Chronic Kidney Disease
• Obesity

Example prompts for Claude

Show me the real-world evidence for Type 2 Diabetes across all dimensions.

What is the geographic distribution of Heart Failure patients in the Northeast between 2020 and 2024?

Compare the longitudinal trend for COPD from 2018 to 2024.

Break down Hypertension prevalence by age and gender only.

Response structure

{ "metadata": { ... }, // source, condition, ICD-10 codes, caveats "summary": { ... }, // total cohort KPIs "by_age": [ ... ], // one row per age group "by_gender": [ ... ], // one row per gender "by_geography": [ ... ], // regions → states "longitudinal": [ ... ] // one row per year }

---

Project layout

rwe_mcp_server/
├── server.py           # MCP server + tool implementation
├── requirements.txt    # Python dependencies (mcp)
└── README.md           # This file