OMOP-aligned ADHD real-world evidence.
An OMOP CDM-aligned ADHD registry. HIPAA Safe Harbor de-identification, FHIR/HL7v2/CSV ingestion, k-anonymity-enforced cohorts for researchers.
Real-world evidence studies live and die on two things: data quality and privacy compliance. Poorly-mapped concepts turn ADHD subtype prevalence into fiction. Sloppy de-identification turns a registry into a lawsuit. Both problems get worse the further data travels from where it was captured.
Alluvion enforces OMOP CDM at ingestion, runs HIPAA Safe Harbor de-identification automatically, and checks every cohort export against a k-anonymity floor. Researchers can't accidentally pull a cohort of two people from a rare-comorbidity slice — the system won't allow it. The safeguards are built in so the science can move faster.
Conditions, drugs and measurements are mapped to the OMOP Common Data Model — SNOMED, RxNorm and LOINC — so cohorts are expressed in the same vocabulary the rest of the OMOP world already uses.
Automated deidentification following the eighteen Safe Harbor identifiers — no cohort is exposed before it's cleared.
No cohort is released below a minimum of 50 patients, with k-anonymity and l-diversity checks applied on top. The floors are enforced in the cohort builder, not left to trust.
Four ingestion paths — FHIR bundles, HL7 v2 messages, CSV rows and plain JSON — so most clinic exports land without a bespoke integration.
Clinics contribute data through one workspace; researchers query the resulting registry through another. Both are running and open to explore — the registry itself is pre-launch, and is not yet accepting patient data.
Clinic-side data entry, contribution tracking and data-quality scoring. Watch a submission arrive, get validated against Kahn's quality dimensions, and move through de-identification into the registry.
Open the capture consoleCohort browser, a criteria builder over the OMOP vocabulary, and structured exports. Every export ships with a manifest recording the privacy thresholds applied, the vocabulary versions used and the licence terms it was released under.
Open the research portalDe-identified real-world evidence — not for clinical decision-making. Research use only.
Alluvion produces aggregate, de-identified research data. It is not a medical device, it is not diagnostic, and nothing it outputs should be used to direct the care of an individual patient. Clinics contributing data remain the data controller for their own patients under their own DUA.
Alluvion is one of the research tools built inside the Synapse Spark Foundation. Everything we make is part of the same mission — understanding neurodivergent minds and supporting the researchers advancing the science.
← Back to the Foundation