Every UK telehealth service runs on one assumption: that the consultation in front of a patient, whether a video call or a multiple-choice questionnaire, actually works as a clinical safety net, not a sales form. Regulators treat the two models as equivalent in principle. In practice, the model you choose shapes what a prescriber can rely on, what gets documented, and how fast a red flag gets caught before a prescription goes out.
Why the consultation is the highest-risk moment in the pathway
Everything else in a telehealth service is heavily audited: the pharmacy premises, the courier chain, the batch records, the complaints log. The consultation is the one step that happens almost entirely inside the patient's own account of themselves, with no independent way to check it in the moment. A prescriber is deciding whether to issue a prescription-only medicine based largely on what someone typed into a form or said on a call.
That is not a reason to avoid online consultations; it is a reason to build them properly. The GPhC's guidance for registered pharmacies providing services at a distance makes the point directly: a pharmacy offering services online has to be satisfied the model can gather enough information to make the same quality of decision a face-to-face encounter would, not a lesser one dressed up as convenient.
Synchronous video versus structured questionnaires: what each is actually good for
A live video or phone consultation lets a prescriber probe an ambiguous answer, pick up on hesitation, and ask the follow-up question a form can't anticipate. It is the stronger model for anything with a wide range of presentations, interacting conditions, or a meaningful risk of the patient not fully understanding what they're being asked.
A well-built structured questionnaire is not a lesser substitute; it is a different tool. It forces every patient through the same decision logic, produces a clean, searchable audit trail, and scales to volumes a clinician-led call never could. Its weakness is that it only works as well as the branching logic behind it, and it depends entirely on the patient answering honestly with no one there to notice if they don't.
What a compliant consultation needs to capture, whichever format it uses
The format matters less than most operators assume. What regulators, and any prescriber worth trusting, actually care about is whether the consultation produced enough of the right information before a decision got made. At minimum that means:
- Confirmed identity, tied to the patient the prescription will actually be issued to
- Relevant medical history, not just a tick-box for "any conditions?"
- Current medications and known allergies, checked for interactions with the medicine requested
- Contraindications specific to that medicine, not a generic list reused across the catalogue
- Any symptom that should stop the pathway and route to a different channel
- Evidence the patient understood the dosing, risks, and what to do if something goes wrong
Miss any of these and the format, however polished, has not actually done its job.
A refusal or escalation is the consultation model working, not a conversion problem to optimise away.
Red flags and escalation: when the pathway has to stop
A consultation model is only as safe as its refusal path. Every build needs a clear, pre-agreed set of answers that stop the process outright: symptoms that need urgent in-person care, an age or circumstance that changes the clinical picture, or a safeguarding concern that needs a different kind of response entirely. Our safeguarding process and how a remote consultation handles Gillick competence for under-18s both sit on top of this same escalation logic.
A declined or escalated consultation is not a failure of the funnel; it is the model working as designed. Operators who track refusal rates as a conversion problem rather than a safety signal are looking at the wrong number.
Documentation and the audit trail regulators actually ask for
When the CQC or GPhC look at an online service, they are not just checking that a consultation happened; they want to see what was asked, how it was answered, and the reasoning that connected those answers to the decision made. A timestamp and a tick saying "consultation complete" is not evidence of anything.
That record needs to survive contact with a complaint, a duty of candour disclosure, or a formal incident review months later, which is a different bar to clear than just logging the transaction. Treating the consultation record as part of the clinical safety case for the software, rather than as an afterthought attached to it, is what tends to separate services that pass inspection comfortably from those that scramble.
A ten-minute questionnaire can be clinically rigorous. A forty-minute video call can be clinically empty. The format tells you nothing about the quality of the decision behind it, only the record of how it was made does.
The failure modes we see most often
Most consultation problems are not dramatic; they are small design choices that compound. The recurring ones:
- A generic questionnaire reused across medicines with very different risk profiles
- No route for a prescriber to ask one clarifying question before approving or refusing outright
- Consultation answers sitting in one system while the dispensing record sits in another, with no easy cross-check
- Identity checks treated as a one-off step rather than something tied to the specific order
- Approvals going out with no documented clinical rationale a reviewer could actually follow later
None of these show up in a demo. They show up in an inspection, a complaint, or an adverse event report, by which point they're expensive to fix.
Where PExpo fits in a consultation flow built this way
PExpo doesn't replace clinical judgement and isn't trying to. What it sits underneath is the layer a consultation decision has to connect to cleanly: identity verification tied to the order, a dispensing record that matches what the prescriber actually approved, and the audit trail a regulator will eventually ask to see. Brands and clinics building on our white-label platform get that connection out of the box rather than having to wire it themselves.
That matters more than it sounds. A consultation model that is clinically sound but hands off to a dispensing process that can't prove what happened is still an exposed service; the two have to hold together end to end.
There is no single correct consultation model for UK telehealth, and regulators have deliberately left room for both video-led and questionnaire-led services to operate well. What they won't tolerate is a consultation that can't show its working: what it asked, how the patient answered, and why a prescriber signed off on the result. Get that part right and the choice between synchronous and asynchronous becomes an operational decision, not a compliance gamble. For a wider look at how the rest of the pathway holds together, see our guide to clinical SOPs for telehealth operators, or talk to us about how PExpo's clinic and white-label paths handle the layer underneath the consultation.
Frequently asked questions
Is a questionnaire-based online consultation as valid as a video call under GPhC rules?
Yes, in principle. The GPhC's guidance on pharmacy services at a distance does not mandate either format; it requires that whichever model is used gathers enough information for a prescriber to make a safe, individualised decision. A well-designed questionnaire can meet that bar, and a poorly designed video call can fail it.
What should happen when a patient's answers raise a red flag mid-consultation?
The pathway should stop and route the patient somewhere more appropriate, rather than letting the consultation complete anyway. That might mean referring to their GP, an urgent care setting, or simply declining to prescribe. The decision and the reasoning behind it should be recorded, not just the outcome.
How long should consultation records be kept, and what should they include?
Records should capture what was asked, how the patient answered, and the clinical reasoning that connected those answers to the prescriber's decision, kept in line with the service's data retention policy and any applicable NHS or ICO record-keeping guidance. A bare log of "consultation completed" is not sufficient on its own.
