Why Case Processing Is So Hard to Scale for a Pharmacovigilance Service Provider
When we talk to pharmacovigilance service providers and CROs, the same problems come up in almost every conversation. The team is good, the clients are happy, and still every new contract means hiring and training more case processors, because each case takes about the same amount of manual work as it did five years ago. These are the problems we hear most often, and what we have learned about each one.
Every client works in a different way
A service provider rarely has two clients with the same set-up. One sponsor runs its own safety database, another submits through EVWEB, a third has a narrative template and coding conventions that nobody else uses. A processor who moves between clients during the day has to switch rules every time, and most QC findings come from that switching: a field filled the way client A wants it in a case for client B, or a narrative written in the wrong template.
The price per case is fixed and the time per case is not
Most case processing contracts are priced per case or as a monthly fee for an agreed volume. The price is set when the contract is signed, while the time a case takes depends on the source: a clean spontaneous report from a healthcare professional takes minutes, a 40-page literature article or a partner email with three products takes much longer. The margin depends on the mix of cases, which the service provider does not control.
Volume peaks arrive faster than new processors
A product launch, a batch of follow-ups from a licence partner or a large literature hit can double the queue in a week. The 15-day and 90-day reporting deadlines stay where they are. A new case processor needs weeks of training on each client’s conventions before working alone, so the peak is usually covered with overtime, and overtime is when errors go up.
Sponsors audit the case history, not just the case
During an audit, the sponsor wants to see how a case got to its final state: which source it came from, who changed each field and when, and why. If that history lives partly in the safety database, partly in email and partly in a tracking spreadsheet, preparing for each audit takes days.
Sponsors ask where the savings from automation go
Sponsors know that automation tools exist for case intake, and some now ask their service provider how much of the work is automated and whether the price per case reflects it. A service provider that adds a tool needs an answer: a lower price per case, a shared target, or more capacity at the same price. It is easier to agree this at the start of a contract than to defend it at renewal.
What to agree with your clients before you use AI on their cases
The marketing authorisation holder keeps responsibility for its pharmacovigilance system, and most PV agreements ask the service provider to report changes to how cases are processed. Tell each sponsor which steps the tool performs, that a qualified processor reviews and approves every case, and how the tool was validated. Sponsors also ask where the data is hosted and who can access it, so have the data processing terms ready before that conversation.
Where Vigintake fits
Vigintake prepares each incoming case for review: it extracts the data, proposes the MedDRA coding and drafts the narrative in each client’s template. A processor checks the case against the source and approves it, and the case goes to the client’s own safety database. Each client keeps its own set-up, and teams that use it spend 80% less time drafting narratives.
If you want to see it with your own material, send us two or three anonymised source documents from different clients and we will process them with you in a 30-minute call. Get in touch or read more about case processing.