Patient safety incident reporting in UK telehealth: what LFPSE means for digital clinics

Most telehealth operators have a Yellow Card process for medicines and a duty of candour policy for serious harm. Fewer have a working answer for the broader category in between: the near-misses, the process failures, the things that went wrong but didn't reach the threshold for either. That gap is what NHS England's Learn from Patient Safety Events service exists to close, and CQC inspectors increasingly ask about it directly.

What LFPSE replaced, and why it exists

Learn from Patient Safety Events (LFPSE) is NHS England's national service for recording patient safety incidents. It replaced the older National Reporting and Learning System (NRLS), which had run since the early 2000s and had become difficult to extend as care moved online and across more provider types. LFPSE is designed to capture incidents from any care setting, including private and independent providers, not just NHS trusts.

For a digital clinic, the point of LFPSE isn't compliance theatre. It's a structured way to notice patterns, a missed allergy flag here, a delayed escalation there, before they compound into something serious. Operators who treat it as a box-ticking exercise tend to miss exactly the signal it's built to surface.

Who has to report, and what counts as a reportable incident

Any provider registered with the CQC is expected to have a functioning incident reporting process as part of demonstrating safe care, and LFPSE is the route NHS England provides for feeding that data into the national picture. A reportable incident is broader than most new operators assume: it isn't only the cases involving actual harm.

• Actual harm to a patient, regardless of severity
• Near misses, where harm was avoided by luck or a last-minute catch
• Process failures that could plausibly have led to harm, such as a missed safety-netting call
• Medication errors that didn't reach the patient, such as a dispensing pick error caught at check
• Systemic issues flagged by more than one clinician, even without a single triggering event

The instinct to only log the dramatic cases is the most common mistake. Near misses are where the useful learning lives, precisely because nobody was harmed and the team can talk about it without defensiveness.

How this differs from Yellow Card and duty of candour

These three obligations get conflated constantly, and they cover different things. Yellow Card reporting, run by the MHRA, is specifically for suspected adverse drug reactions and medical device problems; it's about the medicine or device, not the process around it. Duty of candour is a separate legal and professional obligation that triggers when a patient has come to moderate or severe harm and requires open disclosure to the patient, regardless of whether an LFPSE record is also made.

Patient safety incident reporting through LFPSE sits underneath both, capturing the wider set of events, including the ones that never reach a patient or never involve a medicine at all. A clinic that only has processes for the first two will still have a gap CQC inspectors can find, because most of what actually goes wrong day to day is process-level, not drug-level or harm-level.

Building a workflow clinicians will actually use

The workflow that fails is the one that asks a busy prescriber to fill in a long form during a consultation. The workflow that works separates the moment of noticing from the moment of recording.

1. A single, low-friction channel for any team member to flag a concern, even informally, at the point they notice it
2. A named individual, often the clinical governance lead or superintendent pharmacist, who triages flags within a set number of working days
3. A simple severity categorisation, no harm, low harm, moderate harm, severe harm or death, applied consistently
4. A decision at triage on whether the incident also needs an LFPSE submission, a duty of candour conversation, or both
5. A short written outcome, even for near misses, so the learning doesn't evaporate
6. A recurring review, monthly is typical for a growing clinic, where patterns across incidents get discussed as a team

That last step is the one most operators skip under time pressure, and it's the one that turns a reporting system from an obligation into an actual safety improvement loop.

What CQC inspectors actually look for

Under the Single Assessment Framework, inspectors assessing the 'safe' key question will typically ask to see evidence of a working incident reporting process, not just a policy document describing one. That means: can staff describe how they'd raise a concern, can the provider show recent examples (anonymised or redacted as appropriate), and is there evidence that flagged issues led to a change, not just a record.

A policy that has never generated a single recorded incident in twelve months of live operation reads as a red flag, not a clean record. Inspectors have seen enough digital clinics to know that a genuinely quiet system is rare; more often it means nobody is reporting, which is a different and worse problem.

A genuinely quiet incident log after twelve months of live operation isn't a clean record. It usually means nobody is reporting, which is the worse problem.

Common failure points specific to remote settings

Remote consultation introduces failure modes that a walk-in pharmacy or in-person clinic doesn't have to design around. Recognising these early makes the reporting process more useful, because the team knows what patterns to watch for.

• Safety-netting advice given verbally or in a message that the patient never actually reads
• A patient's stated allergy or condition not carrying across from triage form to prescriber to dispensing pharmacy
• Delayed escalation when a patient's symptoms worsen between consultations, with no clear route back in
• Technology failures, a dropped video call or an unsent message, mistaken for patient disengagement
• Handover gaps between locum prescribers who don't have full visibility of a patient's recent history

None of these are unique to any one platform. They're structural to remote care, which is exactly why they need a reporting route that doesn't depend on any single person remembering to raise them.

How PExpo supports incident reporting in practice

PExpo's dispensing layer sits inside the same clinical governance structure a CQC-registered clinic already runs, so incidents that originate on the dispensing side, a stock substitution, a delivery exception, a labelling query, flow into the clinic's existing reporting process rather than sitting in a separate, disconnected system. That matters for the pattern-spotting this whole process exists to enable; see our guide to writing clinical SOPs for how the two fit together operationally.

We don't file LFPSE reports on a clinic's behalf, that responsibility sits with the CQC-registered provider, but we do make sure the operational data a clinic needs to triage an incident (batch numbers, dispatch timing, who touched an order) is available quickly rather than requiring a support ticket. If you're setting this up alongside a wider compliance build, our Single Assessment Framework guide covers the rest of the 'safe' key question.

Patient safety incident reporting isn't a document you write once and file. It's a habit a clinical team builds, and it only works if raising a near miss feels routine rather than like an admission of failure. Pair it with the processes covered in our guides to duty of candour and safeguarding in remote consultations, and review all three together at your next clinical governance meeting rather than treating them as three separate boxes to tick.

Frequently asked questions