Most medical information enquiries are unremarkable. A pharmacist wants to know whether a tablet can be split. A nurse asks about storage once a vial is opened. A patient read the leaflet twice and still isn’t sure what to do about a missed dose.
None of those is a customer service call, and none is a clinical consultation. They sit in a category of their own, and pharmaceutical companies handle them through a medical information function — sometimes in-house, increasingly through a pharmaceutical contact center that takes first-line volume against approved content.
What makes this work different from general healthcare call handling is not the phone system. It is the rule that an agent may only say what the company’s approved materials already say, and that everything else gets routed rather than answered.
What a medical information call center actually handles
The people who contact a medical information line are mostly healthcare professionals — physicians, pharmacists, nurses — with product questions the label does not obviously settle. Patients and carers make up a smaller share, usually with practical questions about a medicine they are already taking.
The common request types:
Product questions from healthcare professionals. Dosing in a specific situation, administration and preparation, storage and stability, interactions, what the trial data actually showed. These arrive by phone, email and increasingly through web forms.
Patient and carer questions. Missed doses, device or administration technique, what a leaflet passage means in practice.
Product availability and supply. Frequently misrouted to commercial teams when the underlying question is medical.
Requests for materials. A prescribing information copy, a response document, a published reference.
Two things surface on these calls without being asked for: suspected adverse events and product quality complaints. Both have to be recognised and captured regardless of what the caller originally rang about. That is covered below.
MI versus a general healthcare call center
They look similar from outside. They are not the same service, and the differences all point the same way.
What the agent is allowed to say. A general healthcare agent works from a knowledge base someone in operations maintains. An MI agent works from content that medical, regulatory and legal have approved — standard response documents, approved response letters, the prescribing information. If the answer is not in the approved content, the agent does not construct one.
What happens at the edge of the script. A general call center escalates when a caller is unhappy or the question is hard. An MI call center escalates on a defined trigger: the request falls outside approved content, or it needs medical judgement. The escalation is a documented routing step, not a judgement call about difficulty.
What gets recorded. General call handling records the interaction. MI records a case — the enquirer, whether they are a healthcare professional, the product, the question as asked, what was provided, and what was routed onward. That record is what a company relies on if the enquiry is ever revisited.
Who is on the other end of the escalation. In a general operation, a supervisor. In MI, the company’s own medical information specialists.
What the tolerance for improvisation is. A general agent who paraphrases helpfully is doing their job. An MI agent who paraphrases has created an unapproved statement about a medicine.
If you are weighing an MI provider against a general healthcare call center on price, this is the list the difference is hiding in.
Off-label and unsolicited requests
Some requests ask about a use the approved label does not cover. Companies handle these through a defined process for unsolicited requests, and the first-line rule in that process is consistent across the industry: the agent does not answer it.
What the agent does instead is capture it — what was asked, by whom, in what context — and route it to the medical information team under the company’s own procedure. The response, if there is one, comes from that team, in the form the company has approved.
The reason for the strictness is that a company responding freely to off-label questions starts to look like a company promoting off-label use. The distinction is procedural, and it depends on the first-line agent following it without exception.
What to check with a provider: whether the routing step is a documented process with a record, or an instruction given to agents in training. The two are not the same, and only one survives an audit.
Adverse events surface on medical information calls
A caller phoning about dosing may mention, halfway through, that the patient developed a rash. That is an adverse event report arriving on a medical information line, and it is the most common way it happens.
Three things follow, and a provider should be able to state all three without hesitating:
Agents are trained to recognise one. Not to assess it — to notice that something has been described and that the intake procedure now applies.
The required information is captured against a defined field set. An identifiable reporter, a patient, a suspect product, an event. Incomplete reports still get forwarded; the gaps are recorded as gaps.
Reportability is not the agent’s decision, and not the provider’s. Whether an event is serious, expected, or reportable is determined by the company’s pharmacovigilance team. A provider that offers to make that call has misunderstood the arrangement, and that misunderstanding is worth ending the conversation over.
The timeline for getting the information to the safety team should reference the client’s own pharmacovigilance agreement, not a generic service-level number.
What “intake” actually means
Intake gets used loosely. In an MI context it means the first-line capture layer: taking the enquiry, identifying the enquirer, classifying the request, recording it in the client’s case management system, and routing whatever cannot be answered from approved content.
Intake is where most outsourcing arrangements start, and where the volume is. Companies keep response authoring, medical review and reportability determinations in-house, and hand out the layer underneath: answering the phone, working through the routine approved responses, and getting everything else documented and routed accurately.
That split is worth being explicit about when scoping, because it decides how the work is measured. Intake quality is a documentation-accuracy and routing-accuracy question, not an average-handle-time question.
Multilingual and global coverage
A company selling in several markets does not have one medical information line; it has an obligation in every market where it sells. Coverage questions are usually where an in-house team stops scaling.
Language. Enquiries have to be handled in the caller’s language, and the approved content has to exist in that language. Translating an approved response on the call is not an option — that produces an unapproved statement.
Time zones. Enquiries arrive when prescribers are working. A US-hours team covering European markets is a queue by any other name.
Local routing. Escalations go to the affiliate or the team responsible for that market, which means the routing table is part of the setup rather than an afterthought.
Offshore delivery earns its place here specifically: coverage across time zones and languages is expensive to staff onshore, and Philippine delivery runs 40–60% below onshore for the same coverage.
How to evaluate a medical information vendor
Most providers describe the same service. These are the questions that separate them.
Where does your approved content come from, and who can change it? The answer should be: from the client, and nobody at the provider. A vendor that maintains its own response library on your behalf has taken on a role that belongs to medical affairs.
Show me a case record. Redacted, from a comparable account. The completeness of that record is the service.
What happens to a request you cannot answer? You are listening for a documented routing step with a named destination and a record — not “we escalate.”
Who decides whether an adverse event is reportable? The only acceptable answer is: the client’s safety team.
How do you handle a caller who will not identify themselves? A real question with a real procedure behind it, and a good test of whether the provider has actually run this work.
What does your training cover before an agent takes a live call? Product, approved content, AE recognition, and the escalation triggers — at minimum.
Can you work inside our systems? MI cases belong in the client’s case management platform. A provider that wants to run them in its own tooling is creating a reconciliation problem for someone.
If a provider is comfortable with all seven, the rest of the conversation is about coverage and cost. If they are vague on two or three, the price is not the thing to negotiate.
Closing
Outsourcing medical information intake is not a decision about headcount. It is a decision about which layer of a regulated process sits outside the company, and how tightly its boundaries are drawn.
The pharmaceutical contact center page sets out what Magellan handles across medical information, patient support programmes and adverse event intake — including a plain statement of where the scope ends — along with the systems the team works inside and how onboarding runs.
