First Contact with a Number: A Laboratory Playbook for Direct-to-Patient Results

Direct-to-patient release has moved from pilot project to routine expectation. In England, the NHS App and many local patient portals now show blood test and pathology results, often within hours of authorisation. The laboratory result itself has not changed. What has changed is who reads it first, and in what context. A potassium of 6.1 mmol/L, a haemoglobin of 101 g/L or a TSH just outside the reference interval may reach a patient's phone before anyone has explained whether the value is urgent, expected or simply a laboratory artefact.

For UK laboratory teams in 2026, this is not a public-relations problem. It is a diagnostic-safety and communications problem that sits squarely inside the pathology workflow. The practical questions are unglamorous but consequential: what does the report say, when is it released, who is alerted, and what happens when the number is ambiguous?

Why the first reader matters

A laboratory report is written for a clinical reader who knows the patient, the indication and the pre-analytical history. A patient sees the same report without that scaffolding. They may not know that a low ferritin in an inflammatory state is hard to interpret, that a mildly raised creatinine can reflect dehydration, or that a tumour marker is not a screening test. They may search the number online and arrive at a conclusion the laboratory never intended.

That does not argue against access. Patients have a legitimate interest in their records, and direct access can support shared decision-making, adherence and trust. But access without context converts an interpretive document into a raw data feed. Laboratories should treat the patient-facing view as a distinct output of the diagnostic process, with its own wording, timing and governance.

The interpretive comment is not a disclaimer

Many laboratory reports already carry comments, but these are often written for GPs and specialists. Phrases such as 'correlate clinically' or 'repeat if indicated' are sensible shorthand between professionals and close to useless for a patient reading at home. A patient-facing comment should do three things: say what the result means in plain language, say what it does not mean, and say what happens next.

For example, a report might note that a mildly raised TSH with a normal free T4 usually requires a repeat test rather than immediate treatment, and that the patient should wait to hear from their GP. A low haemoglobin should not be labelled simply 'abnormal'; the comment should distinguish a probable iron-deficiency picture from a result that needs urgent clinical review. Wording should be version-controlled, approved by clinical leads and reviewed when reference intervals or assay platforms change. It should never offer a diagnosis or a treatment instruction.

There is a subtle risk here. Standardised comments can become so generic that they add words without adding meaning. Laboratories should test patient-facing text with actual patients and with primary care colleagues, not only with laboratory staff. The question is not whether the comment is scientifically correct, but whether it prevents a reasonable person from drawing an unreasonable conclusion.

Timing is a clinical decision

Release timing is often treated as a portal configuration issue. It is better understood as a clinical decision. Routine results with clear interpretive comments may be safe to release immediately after authorisation. Results that are markedly abnormal, or that depend on a clinical conversation, may need a short hold while the responsible clinician is notified. Critical results should follow the existing urgent-notification pathway, not the portal.

The difficulty is that 'abnormal' is not a single category. A raised ALT may be urgent in one patient and unsurprising in another. A positive pregnancy test, a genetic variant or a sexually transmitted infection result carries different communication needs from a lipid profile. Laboratories and clinical teams should agree release rules by test group and by clinical setting, with explicit exceptions for results that require same-day discussion. These rules should be documented, auditable and reviewed after incidents.

Escalation when the number is ambiguous

Ambiguity is the hardest case. A result may be technically valid but clinically unclear: an unexpected positive, a borderline serology, a variant of uncertain significance, or a result that conflicts with a previous sample. In a clinician-only workflow, the laboratory rings the GP and the uncertainty is discussed. In a direct-release workflow, the patient may see the result while that conversation is still happening.

A practical escalation pathway should identify who owns the result once it leaves the laboratory information system, how quickly a clinician must be alerted, and what the patient-facing view shows in the interim. Some services use a 'result available, interpretation pending' state. Others hold selected categories until a clinician has acknowledged them. There is no single correct model, but there must be a named owner and a tested route. The portal should not become the escalation route for a critical result.

A minimum release specification for laboratories

A defensible direct-release specification need not be complex. It should state which tests are released immediately, which are held, and why. It should define patient-facing comment standards and require sign-off from clinical and patient representatives. It should include an equality impact assessment, because not all patients have smartphone access or health literacy. It should preserve the full laboratory record, including comments and amendment history, so that a later clinician can see what the patient saw. And it should include a route for patients to ask questions that does not depend on the laboratory's main switchboard.

The bench-to-bedside view

Pathology has spent years improving analytical quality, turnaround times and interoperability. Direct-to-patient release adds a new quality dimension: the first-contact experience of a result. Laboratories cannot control what a patient does with a number, but they can control the context in which it arrives. In 2026, that context should be designed as carefully as the assay itself.