Platform / Case processing
Case processing that runs End to End
From the first email to the regulator’s acknowledgement, every eligible case flows through one automated path, and every case that needs judgement reaches the right reviewer, fully prepared.
Zero-touch eligible, routed by your rules
Every action, human and AI, logged
Everything case processing needs, in one place
Multi-channel intake
Safety reports arrive from everywhere: inboxes, call centres, partners, patient programmes and literature. They all land in one queue, classified and ready to process.
- Email, PDF, scanned forms, E2B files and web forms
- Call-centre, patient support programme and partner feeds
- Follow-ups matched to existing cases automatically
- Every source document kept on the case
Medical information inboxNew spontaneous caseEmailPartner [Name] · E2B (R3)Partner caseE2BPatient support programmeNew casePSPCall centre · transcriptNew caseCallLiterature module hand-offLiterature caseLitReply from reporterFollow-up · case [ID]EmailExample queue. Channels are configured during setup.
Extraction with sources
Patient, reporter, product, event and timeline details are extracted from every document, and each value is linked to the exact place it came from.
- Structured fields filled from forms, emails and narratives
- Every value linked to its page, line or box
- Confidence shown, with low-confidence fields sent for review
- Powered by Document Intelligence
Coding
Events, indications and medical history are coded to MedDRA and products to WHO Drug, using your coding conventions, with confidence and alternatives shown to the reviewer.
- Verbatim terms coded with the full hierarchy visible
- Your synonym lists and conventions applied
- Low-confidence terms routed to a coder
- Powered by Pharmacovigilance Intelligence
Example terms and confidence scores.
Triage and duplicates
Each case is checked for validity, seriousness and expectedness under your SOP rules, prioritised, and compared against existing cases to catch duplicates across every source.
- Four minimum criteria checked for a valid ICSR
- Seriousness and expectedness assessed by your rules
- Priority set from the reporting clock and risk
- Duplicates found across spontaneous, partner and literature sources
Medical review
Cases that need medical judgement reach a safety physician with the work already done: source-linked data, coded terms, proposed assessments and a draft narrative, all in one review screen.
- Case summary, source documents and coded data side by side
- Proposed seriousness, expectedness and causality evidence shown for review
- Queries and follow-up requests raised from the same screen
- Every medical decision attributed, time-stamped and signed
Seriousness classification
Every case is checked against the internationally agreed seriousness criteria, and against important medical event and designated medical event lists, so serious cases are never missed and always reported on the right clock.
- Death, life-threatening, hospitalisation, disability, congenital anomaly and other medically important conditions
- Important medical event and designated medical event lists applied
- The source text behind each criterion shown
- Final classification confirmed by your reviewer
Expectedness and listedness
Each event is checked against the reference safety information that was in force on the relevant date, such as the company core safety information or the investigator’s brochure.
- Reference safety information held and versioned per product
- The version in force on the right date used automatically
- Listed and unlisted events shown with the matching text
- Unexpected serious cases flagged for expedited reporting
| Event | Reference | Result |
|---|---|---|
| Headache | CCSI v[__] · section 4.8 | Listed |
| Vomiting | CCSI v[__] · section 4.8 | Listed |
| Pancreatitis | Not in CCSI v[__] | Unlisted |
Causality identification
Causality is a medical judgement. vGenix Ai assembles the evidence a physician needs, using the factors common causality methods rely on, and records the reporter’s assessment. The company assessment stays with your experts.
- Time to onset and plausibility of the timing
- Dechallenge and rechallenge information
- Alternative explanations: disease, history, other medicines
- Known reactions for the product and class
- Support for your chosen method, such as WHO-UMC
Narrative drafting
A case narrative is drafted from the coded data in your house style, so every narrative follows the same structure and matches the data exactly.
- Drafted from the structured case, not from scratch
- Your narrative template and house style
- Every statement traceable to the case data
- Reviewed and edited by your team before approval
QC built in
Completeness and consistency checks run while the case is processed, not after it, so errors are fixed before QC review rather than found by it.
- Required fields checked against your SOP
- Dates, doses and outcomes checked for consistency
- Narrative checked against the coded data
- Remaining issues listed for the QC reviewer
Zero-touch and expert lanes
Cases that meet the eligibility rules you define flow from intake to submission with no manual touch. Cases that need judgement go straight to the right expert, fully prepared.
- Eligibility rules agreed with you and validated before go-live
- Eligible cases processed from intake to acknowledgement automatically
- Serious or unexpected cases routed to medical review
- Routing decisions recorded in the audit trail
Clock tracking
Day zero is set the moment a case qualifies, and every case is tracked against its 15- or 90-day clock, with cases at risk escalated in time to act.
- Day zero set from receipt of minimum criteria
- Clocks per market and case type
- Restarts handled for significant follow-up
- At-risk cases escalated before the deadline
Four steps, one continuous flow
Intake
Any channel, any format, any language
Triage & extract
Validity, seriousness, source-linked fields
Code & QC
MedDRA, WHO Drug, consistency checks
Submit & ACK
E2B (R3) out, acknowledgement tracked
Zero-touch lane
Eligible cases flow from intake to E2B (R3) submission and acknowledgement with no manual touch. Eligibility rules are agreed with you and validated before go-live.
Expert lane
Cases that need judgement, such as serious or unexpected events, are routed automatically to the right reviewer with everything prepared.
AI does the work. Your experts decide.
- Intake from every channel
- Extraction and translation
- MedDRA and WHO Drug coding
- Duplicate detection
- Narrative drafting
- QC checks
in the
loop
- Medical review of serious cases
- Causality assessment
- Final QC approval
- Unexpected-event escalation
- Follow-up decisions
- Sign-off before submission
vGenix Ai vs other approaches
How case processing works today in most teams, with point tools, and with vGenix Ai.
| Area | Manual process | Point-tool automation | vGenix Ai |
|---|---|---|---|
| Intake | Inboxes and forms checked by hand | OCR tool per channel | Every channel in one queue, classified automatically |
| Data entry | Re-keyed into the safety database | Extracted, then re-keyed between tools | Extracted once, source-linked, no re-keying |
| Coding | Manual MedDRA and WHODrug lookup | Suggestions without your conventions | Coded to your conventions, with confidence and reasoning |
| Triage | Reviewer judgement case by case | Simple keyword rules | Validity, seriousness and expectedness by your SOP rules |
| Duplicates | Spotted if someone remembers | Checked within one source | Checked across every source |
| Routine cases | Every case touched several times | Still touched between tools | Zero-touch for eligible cases |
| Traceability | Notes and spreadsheets | Gaps between tools | Every value and decision in one audit trail |
| Deadlines | Tracked in trackers | Partly tracked | Every case on its clock from day zero |
“Point-tool automation” describes separate tools that each automate one step. It does not refer to any named vendor.
What changes, and how we measure it
Every engagement starts with your baseline. These are the measures we track for case processing.
Illustrative scenario: a mid-size pharma company handling about 40,000 ICSRs a year across 3 regions. These figures are for discussion, not client results. In every engagement we agree baselines with you before go-live and measure results from platform data.
- Case cycle time
- Touch time per case
- Share of eligible cases processed zero-touch
- On-time submission rate
- Data-entry accuracy
Part of one platform
Case processing shares its data model, rules and audit trail with every other module, so nothing is re-keyed between them.
Intelligent safety should be one connected operation
Automate the work. Keep the decisions with your experts. Keep the evidence traceable. See it on your own data in six weeks.
6-week pilot on literature monitoring
- Baselines agreed up front
- Configured to your SOPs
- Measured on your own data
- Validation evidence from day one