PopHIVE MCP Server

MCP server exposing near real-time public health surveillance data from Yale's PopHIVE platform, plus HHS/CDC datasets, for analysis in AI agents.

Healthcare & Life Sciences by Cicatriiz (community) None active

GitHub Docs

Overview

PopHIVE MCP Server is a community-built Model Context Protocol server that wraps the Yale School of Public Health's PopHIVE (Population Health Information Visual Explorer) platform along with additional CDC and HHS open datasets. It gives AI agents structured access to near real-time, de-identified surveillance data covering immunizations, respiratory illness, chronic disease, hospital capacity, injuries and overdoses, and youth mental health.

The server aggregates data from authoritative sources including the CDC National Immunization Survey, Epic Cosmos EHR Network, CDC Laboratory Surveillance (NREVSS), CDC Wastewater Surveillance (NWWS), Google Health Trends, HealthData.gov, and data.cdc.gov. It exposes five tools for filtering, comparing states, time series analysis, dataset discovery, and search, plus four pre-built MCP prompt templates for common epidemiological analyses such as detecting respiratory surges or identifying immunization coverage gaps.

The project is MIT licensed, requires no authentication (all underlying data is public), and ships as a Node.js server with a Desktop Extension (.dxt) bundle for one-click install into Claude Desktop. It is maintained by an independent developer (Cicatriiz) rather than Yale or the CDC, and was featured on Claude as a reference public health MCP integration.

Tools

Tool	Description
`filter_data`	Filter a PopHIVE dataset by state, date range, demographics, and conditions.
`compare_states`	Compare a health metric across multiple states for a given dataset and time period.
`time_series_analysis`	Analyze temporal trends in a dataset with aggregation options.
`get_available_datasets`	List all available datasets with metadata and optional sample data.
`search_health_data`	Search across datasets for conditions or keywords.
`immunization_gaps (prompt)`	Prompt template that analyzes vaccination coverage gaps by demographics.
`respiratory_surge_detection (prompt)`	Prompt template to detect and characterize respiratory disease surges.
`chronic_disease_trends (prompt)`	Prompt template to analyze chronic disease prevalence trends over time.
`multi_source_analysis (prompt)`	Prompt template that integrates ED, lab, wastewater, and search data for a unified view.

Setup Guide

Prerequisites

Node.js 18 or later
npm or yarn
An MCP-compatible client (Claude Desktop, Cursor, etc.)
No API key or authentication required

Option A: Desktop Extension (recommended)

Download the latest .dxt file from the GitHub releases page.
Double-click the file to open it with Claude Desktop.
Click "Install" in the dialog.

Option B: Manual install from source

git clone https://github.com/Cicatriiz/pophive-mcp-server.git
cd pophive-mcp-server
npm install
npm test
npm start

MCP client config

Add this to your Claude Desktop or Cursor MCP config file:

{
  "mcpServers": {
    "pophive": {
      "command": "node",
      "args": ["server/index.js"],
      "cwd": "/path/to/pophive-mcp-server"
    }
  }
}

Optional environment variables

Variable	Default	Notes
`DATA_CACHE_DIR`	`./data`	Local cache directory for downloaded datasets
`UPDATE_FREQUENCY`	`daily`	One of `hourly`, `daily`, `weekly`
`NODE_ENV`	`development`	Use `production` for deployments

Use Cases

Detect early respiratory disease surges by cross-referencing ED visits, lab surveillance (NREVSS), wastewater signals, and Google Health Trends in a single agent workflow.
Identify state-level childhood immunization coverage gaps using CDC NIS and Epic Cosmos data, then generate targeted outreach summaries.
Compare chronic disease prevalence (obesity, diabetes) across selected states or demographics for grant applications and public health reports.
Monitor near real-time hospital capacity from HealthData.gov to flag stress in specific regions.
Pull injury, overdose, and youth mental health ED metrics from data.cdc.gov for journalism, research, or policy briefings.

Example Prompts

"Compare RSV ED visits per 100k in California, Texas, and New York for the last 12 weeks."
"Show me states with the largest drop in MMR coverage among 24-month-olds in the most recent NIS data."
"Has there been a respiratory illness surge in the Northeast in the past month? Combine NREVSS, wastewater, and Google Health Trends."
"Plot weekly diabetes prevalence rates from Epic Cosmos for Florida from 2022 to 2025."
"List all available PopHIVE datasets and tell me which ones support state-level filtering."

Pros

No authentication or API key required; all underlying PopHIVE, CDC, and HHS data is public.
Broad dataset coverage (11 datasets) spanning immunizations, respiratory illness, chronic disease, hospital capacity, injuries/overdoses, and youth mental health.
Ships as a Desktop Extension (.dxt) for one-click install in Claude Desktop.
Includes pre-built MCP prompt templates for common epidemiological analyses.

Limitations

Community-maintained by an independent developer, not officially endorsed by Yale, the CDC, or HHS.
Geographic granularity is limited: respiratory_lab and injury_overdose are national only, and wastewater data is regional rather than state-level.
Update frequencies vary widely by dataset (some annual, like NIS immunizations), so "near real-time" does not apply uniformly.

Alternatives

CDC Open Data via generic HTTP/Socrata MCP servers for direct queries against data.cdc.gov.
HealthData.gov MCP integrations for hospital capacity and utilization datasets only.
Custom MCP wrappers over the Epic Cosmos or PubMed APIs for clinical research workflows.