Real-world evidence, from your own data, without ever exporting it.
Marketed-drug safety signals, label extensions, comparative effectiveness: the answers live in real-world data you already hold. Polnor turns that data into evidence inside your own cloud, on an HDS-certified host, with a pipeline where almost every step runs itself.
Is a marketed drug associated with an adverse event?
A pharmacovigilance question, answered on real-world data. You define exposed and comparator cohorts, adjust for confounders, and read a risk estimate, without a single row leaving your cloud. Here is the path, and how little of it is manual.
| cohort | n | event rate | HR (adj.) |
|---|---|---|---|
| exposed | 41,802 | 3.1% | 1.28 |
| comparator | 41,802 | 2.4% | ref |
Illustrative interface and figures, shown to explain the workflow. Not results from a Polnor customer.
Everything up to the cohort is automated.
By the time you write that query, the hard part is done: the platform has already ingested, de-identified, standardized and quality-checked your data. You start where the science starts.
Real-world data in, at the FHIR standard
Claims, EHR extracts and lab feeds are pulled and normalized to FHIR R4 automatically, one typed table per resource. No bespoke ETL per source.
| resource | rows | status |
|---|---|---|
| DrugExposure | 8,912,004 | ✓ auto |
| ConditionOccurrence | 4,201,550 | ✓ auto |
| Observation | 12,004,933 | ✓ auto |
GDPR de-identification and OMOP mapping, hands-off
Direct identifiers are masked and dates shifted by preset, then the data is converted to the OMOP CDM with OHDSI concept mapping and clinical quality control. The comparator matching you rely on runs on standardized concepts, not free text.
| source | → OMOP concept | std |
|---|---|---|
| atc:X | ingredient X | S |
| icd10:I26 | Pulmonary embolism | S |
| loinc:718-7 | Hemoglobin | S |
Cohorts, adjustment and models, where the science is
This is the part your epidemiologists own: define exposed and comparator cohorts, adjust for confounders, and, if you need it, train and track a model with MLflow. Everything runs on compute next to the data, and nothing is exported.
| model | AUC | endpoint |
|---|---|---|
| ae-risk-drugX | 0.83 | /v1/risk |
| propensity-match | 0.89 | /v1/ps |
The platform does the plumbing. Your team does the science.
The line is deliberate: the repetitive, error-prone, compliance-heavy work runs itself, so your scientists spend their time on judgment, not preparation.
Evidence you can defend, with a trail behind it.
Every access to patient data is logged, PHI is classified, and export follows EHDS-aligned manifests. Because the data never leaves your HDS-certified cloud and we never train our models on it, the provenance of your evidence is unambiguous.
From raw RWD to a defensible estimate, in your cloud.
Book a 30-minute demo. We run the automated pipeline on a case that looks like yours, and show you exactly where the automation ends and the science begins.