Practice Extract
Realistic, fully-synthetic patient cohorts in the standard UK primary-care research format. Delivered as ten linked tables covering demographics, clinical events, prescriptions, tests, and consultations.
Rhysora Practice Extract is synthetic data generated from published sources. It contains no real patient data and is not clinical data. It must not be used to inform decisions about real patients or as real-world evidence where regulators require real data.
How Rhysora compares to CPRD synthetic data
CPRD's synthetic datasets are derived from anonymised real patient data. Access requires a non-negotiable Data Sharing Agreement with MHRA — typically weeks of procurement before the first download.
Rhysora is generated from published reference sources alone. No real patient data is used at any stage, so there is nothing to anonymise and no Data Sharing Agreement to sign. For methodology development, EHR testing, training, and early research prototyping, this means a licence conversation instead of a regulatory process.
Rhysora is not a replacement for CPRD where real-data evidence is needed. For regulatory submissions that require real-world data, CPRD remains the appropriate route.
What you get
A complete synthetic patient population sized to your needs. Each patient has a realistic medical history: demographics and registration details, consultations over time, prescriptions with BNF codes, lab tests with reference ranges and results, referrals, immunisations, clinical notes with SNOMED coding, and family history where relevant.
Delivered in the standard UK primary-care research format used by CPRD and similar databases. Ten linked tables covering demographics, clinical events, prescriptions, tests, and consultations.
You choose
- Cohort size. From ten patients to hundreds of thousands.
- UK location. Any region. Demographics and prescribing patterns reflect local profiles.
- Reference date. Any study period. Patients age realistically around it.
- Data shape. Clean research-ready or realistic real-world-EHR messiness.
How it's delivered
Tab-separated files in the standard schema. Ready to load into R, Stata, SAS, Python, or any database. No proprietary viewer required.
Who uses it
Academic research
Prototype study designs before committing to real-data access applications. Validate analysis pipelines on known-shape data.
Pharma HEOR
Build evidence synthesis pipelines and analytic frameworks. Develop and stress-test methodology ahead of real-data deployment.
EHR vendors
Test products against realistic messy data. Run demonstrations and training without patient-data access constraints.
Teaching and training
Educational datasets for epidemiology, biostatistics, and health informatics courses. No approvals, no risk.
Why Rhysora
Every value in every record traces back to a published source. NICE guidelines, British National Formulary, ONS data, NHS datasets, and the pharmacoepidemiology literature.
Every run is reproducible. The same settings produce the exact same dataset on any computer. Findings built on Rhysora data can be independently verified.
Outputs are continuously reviewed against published standards. Read more about our methodology →
Ready to talk?
We respond personally to every enquiry. Tell us about your use case and we'll scope a pilot.
Get in touch →