Export Formats

Export Formats

Synthea can export synthetic patient data in three formats: FHIR R4, CSV, and JSON. Each is controlled independently in synthea.properties.

FHIR R4 (default)

FHIR R4 is enabled by default. Each patient is exported as a transaction Bundle containing all their clinical resources.

Enable

exporter.fhir.export = true

Output structure

output/fhir/
├── hospitalInformation<timestamp>.json     # Organization / Location resources
├── practitionerInformation<timestamp>.json # Practitioner resources
├── Ada_Howell_<uuid>.json                  # One Bundle per patient
└── ...

Bundle contents

Each patient bundle includes:

Resource	Description
Patient	Demographics, identifiers, address
Encounter	All clinical visits
Condition	Active and historical diagnoses
MedicationRequest	Prescriptions
Observation	Labs, vitals, social history
Procedure	Surgical and clinical procedures
Immunization	Vaccination history
AllergyIntolerance	Known allergies
CarePlan	Active care plans
DiagnosticReport	Lab and imaging reports
ImagingStudy	Radiology studies
Claim	Insurance claims
ExplanationOfBenefit	EOB records
Coverage	Insurance coverage periods
Organization	Providers and payers
Practitioner	Clinicians

POST to a FHIR server

exporter.fhir.server_url = https://fhir.example.com/r4
exporter.fhir.transaction_bundle = true

Python: load a bundle

import json
from pathlib import Path

bundle_path = Path("output/fhir/Jane_Doe_abc123.json")
bundle = json.loads(bundle_path.read_text())

for entry in bundle["entry"]:
    rt = entry["resource"]["resourceType"]
    print(rt)

---

CSV

CSV export produces one table per resource type — useful for database ingestion and analytics.

Enable

exporter.csv.export = true

Tables generated

File	Contents
patients.csv	One row per patient
encounters.csv	All encounters
conditions.csv	All diagnoses
medications.csv	All medication orders
observations.csv	Labs and vitals
procedures.csv	All procedures
immunizations.csv	Vaccines
allergies.csv	Allergy records
careplans.csv	Care plan entries
claims.csv	Insurance claims
claims_transactions.csv	Line items
payers.csv	Payer / insurance info
providers.csv	Provider info
organizations.csv	Facility info
devices.csv	Medical device records

Load into pandas

import pandas as pd

patients = pd.read_csv("output/csv/patients.csv")
conditions = pd.read_csv("output/csv/conditions.csv")

diabetic = conditions[conditions["DESCRIPTION"].str.contains("diabetes", case=False)]
diabetic_ids = diabetic["PATIENT"].unique()

diabetic_patients = patients[patients["Id"].isin(diabetic_ids)]
print(diabetic_patients[["FIRST", "LAST", "BIRTHDATE", "GENDER"]].head())

---

JSON (raw)

Raw JSON export writes the internal Synthea patient object, before FHIR mapping. Useful for debugging modules or building custom exporters.

Enable

exporter.json.export = true

Output: output/json/<PatientName>_<uuid>.json

---

Controlling Output

# Where files are written
exporter.baseDirectory = ./output/

# Use UUID filenames (avoids special characters in patient names)
exporter.use_uuid_filenames = false

# Append to existing CSV files (false = overwrite each run)
exporter.append_mode = false

# Skip living patients (emit deceased only)
exporter.only_living = false

Enabling All Three Formats

uv run synthea -p 100 -c <(cat <<EOF
exporter.fhir.export = true
exporter.csv.export = true
exporter.json.export = true
exporter.baseDirectory = ./output/
EOF
)