Medical Affairs · Field Medical · Medical Information
Answer the next HCP question with real-world evidence — not a quarterly study.
Medical teams self-serve characterisation, treatment-pattern and comparative questions across connected biobanks. Authored against OMOP. Aggregate-only by design. Citable artefact for every response.
MSL / FIELD MEDICAL
Evidence for the next meeting, not the next quarter
MEDICAL INFORMATION
SRLs and off-label responses grounded in RWE
SCIENTIFIC COMMUNICATIONS
Congress-ready analyses with a citable artefact
RWE / MEDICAL EVIDENCE
Scale expert capacity across the whole function
01 · The shift
Medical Affairs is expected to speak in evidence. The infrastructure hasn't kept up.
THE OLD CADENCE
- HCP asks about a subgroup. MSL routes to the RWE team.
- RWE team queues it with two others. Scope call in three weeks.
- CRO engaged, CDM decided, SOW signed. Six-to-twelve months pass.
- Answer arrives. Congress is over. The question has moved.
WITH UNISON
- Medical team opens a protocol template or asks in natural language.
- Protocol compiles to UQL and fans out across connected biobanks.
- Aggregate-only results return with provenance. Nothing leaves the custodian.
- Artefact is citable. The response goes back to the HCP the same week.
02 · What Medical Affairs can do, self-serve
Five workflows, one platform.
DISEASE & TREATMENT LANDSCAPE
Characterise the patient population
Prevalence, demographics, comorbidity patterns, line-of-therapy sequencing. The background every scientific narrative starts from.
OUTPUT
Cohort characterisation · treatment-sequence tables · standardised reportCOMPARATIVE EFFECTIVENESS
Real-world outcomes vs comparator
Time-to-discontinuation, switch rates, adherence, HCRU, outcomes proxies. Pre-specified, matched, reproducible.
OUTPUT
KM curves · PS-matched comparisons · effect estimates with CIsUNMET NEED & GAPS IN CARE
Show where patients fall through
Guideline-concordance rates, time-to-diagnosis, under-treated subgroups. Evidence for advisory boards and payer conversations.
OUTPUT
Funnel analysis · gap quantification · subgroup breakdownsMEDICAL INFORMATION
SRLs and off-label enquiries
Turn an ad-hoc enquiry into a standing analysis. Re-run the same query when the data updates, not when the letter does.
OUTPUT
Cohort counts · event rates · standardised response with citationCONGRESS & PUBLICATION
Hypothesis-to-abstract in weeks
Explore, refine, lock. When the abstract submits, the UQL artefact is the methods section. When a reviewer asks, the re-run is a click.
OUTPUT
Reproducible analysis · methods artefact · audit trail03 · A workflow built for medical teams
From question to citable answer, in the hands of the people closest to the science.
01
Ask
Natural language or a protocol template chosen by the medical team — not a SQL request to the data function.
02
Compile
Unison compiles the protocol to UQL — a deterministic, aggregate-only query contract.
03
Federate
UQL fans out across connected biobanks. Data never leaves the custodian; only aggregate results return.
04
Inspect
Per-source results are visible. Heterogeneity is something you can see, not a black box.
05
Cite
Every result is backed by a replayable UQL artefact — citable in a response, an SRL, an abstract.
A DAY IN MEDICAL AFFAIRS
"What does the treatment pattern look like for patients with renal impairment?"
A field medical director takes the question after a KOL meeting. By the time they're back at their laptop, the answer is queued up and waiting.
Scoped
protocol template, refined in place
Executed
federated across 3 biobanks
Governed
aggregate-only, provenance logged
Delivered
SRL response with citable artefact
# unison · medical-affairs workspace
> "Treatment sequences in patients with moderate-to-severe renal impairment, 12m post-index."
→ Template: treatment-sequence · compiled to UQL
→ Federated across 3 biobanks · aggregate-only
Cohort n = 38,214
Line 1 · Agent A 42.6%
Line 1 · Agent B 29.1%
Line 1 · Other 28.3%
Median time to switch 7.4 mo (IQR 3.9–13.2)
# citable artefact · uql://query/ma-214f
04 · Built to pass Compliance, Legal and IT
The safeguards Medical Affairs needs — not bolted on, built into the query language.
Aggregate-only by design
UQL cannot express a patient-level query. Exfiltration is structurally impossible — not merely prohibited by policy.
Promotional-compliance aware
Separate workspaces for on-label and off-label enquiries. Decisions logged, reviewable, reversible.
Custodian-respecting
The analysis travels to the data. Custodians keep control. Every mapping decision is attributable to a person and a moment.
Standards:· OMOP CDM-native· Cyber Essentials Plus· CFR 21 Part 11-ready· EHDS-aligned· Aggregate-only IR
05 · What changes for the function
From "we'll get back to you" to "here's what the data says."
RESPONSE LATENCY
weeks → days
HCP enquiries answered inside the decision window.
TEAM LEVERAGE
1 RWE analyst → 10 medical teams
Expert capacity scales across the whole function.
EVIDENCE SHELF-LIFE
one-off study → living query
Re-run when the data updates, not when the letter does.
Scope a pilot with your medical team in 2 weeks