synthea

Generate and load synthetic FHIR data from PheKB phenotypes using Synthea.

1|Updated Jan 12, 2026

Search Tags:#qa #healthcare #test-data #data-generation #fhir #synthea #phekb

Ready to run

npx skills add https://github.com/feordin/llm-fhir-query-eval --skill synthea

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: synthea
Source: https://github.com/feordin/llm-fhir-query-eval/tree/main/.claude/skills/synthea
Command: npx skills add https://github.com/feordin/llm-fhir-query-eval --skill synthea

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Generate and load synthetic FHIR data from PheKB phenotypes using Synthea.

Core Features & Use Cases

End-to-end phenotype data generation: create Synthea modules, generate synthetic patients, and load data into a FHIR server.
Reusable templates: leverage GMF-based module templates and phenotype data to reproduce synthetic cohorts for multiple phenotypes.
QA and testing readiness: produce distinct positive and control cohorts to validate query accuracy and system behavior.

Quick Start

Create modules with /synthea create-module <phenotype>, then run /synthea full <phenotype> to generate and load data.

Frequently Asked Questions about synthea

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate synthetic FHIR data from PheKB phenotypes for testing?▼

Generate synthetic FHIR data from PheKB phenotypes by running the full pipeline command, which creates Synthea modules, produces synthetic patients, and loads them directly into a FHIR server for testing.

What is the best way to create distinct positive and control cohorts for QA validation?▼

Create distinct positive and control cohorts for QA validation by generating synthetic patient data from phenotype modules. This approach produces separate cohorts to validate query accuracy and verify system behavior across test environments.

Do I need a running FHIR server to use synthetic patient data generation workflows?▼

Yes, a running FHIR server at http://localhost:5826/r4 is required to load the generated synthetic patient data. The pipeline commands expect this endpoint to be active to successfully import the synthetic FHIR records.

How do I reproduce synthetic patient cohorts for multiple phenotypes?▼

Reproduce synthetic patient cohorts for multiple phenotypes by leveraging reusable GMF-based module templates and existing phenotype data. This approach consistently regenerates synthetic cohorts across different testing scenarios.

Where should I place existing Synthea modules and phenotype data for the generation pipeline?▼

Place existing Synthea modules under synthea/modules/custom and phenotype data under test-cases/phekb. These specific directory locations are required for the pipeline to correctly read and process the generation inputs.

What commands are available to manage the synthetic FHIR data generation pipeline?▼

The pipeline provides create-module, generate, load, and full commands to manage the synthetic FHIR data workflow. These commands handle everything from phenotype module creation to FHIR server loading end-to-end.