Doctor’s Passport

Find your next topic

Explore the current textbook

Available drafts · Clinical review pending
Membership
Educational draft · awaiting clinical reviewThe full textbook explains uncertainty but does not replace live national or local guidance, specialist advice, or current prescribing information.
Full textbookMLAMSRAMRCP

A structured approach to laboratory results

Interpret laboratory data through identity, specimen quality, method, reference context, biological variation and clinical probability, while acting safely on critical and changing results.

Saved on this device
Open the sections you need. The overview is shown first.
01Principles and purposeThe professional or clinical skill and the decisions it supports.

A laboratory value is the end product of a chain: selecting a test, preparing and identifying the patient, collecting the correct specimen, transporting it, performing an assay and reporting it. Error or variation at any stage can change the number. Interpretation therefore begins with provenance. A potassium result from a haemolysed sample, a glucose drawn from an infusion line and a coagulation sample with an incorrect fill are different evidence from technically valid specimens.

The reference interval usually contains a defined proportion of results from a reference population. By construction, some healthy people fall outside it, and a value inside it can be abnormal for a particular patient or decision. Age, sex, pregnancy, time of day, posture, diet, exercise, medicines, renal function and analytical method can shift expected values. Clinical decision limits, such as a treatment threshold validated for a particular outcome, must not be casually substituted for a laboratory range.

Pattern recognition is stronger than flag counting. Ask whether related results fit one mechanism, whether the magnitude is plausible, and whether the time course matches physiology. For example, a new large change across several linked analytes may represent real organ dysfunction; a single implausible value amid stable linked measures may indicate collection or transcription error. Delta checks and previous baselines help, but a stable abnormality can still be clinically important.

Uncertainty should change behaviour. Analytical imprecision, biological variation and sampling conditions mean small movements near a threshold may not represent real change. Conversely, a rapidly evolving illness may justify repeat measurement sooner than routine minimum intervals. Repeat testing should have a stated question: confirm an unexpected result, establish trajectory, assess response or detect toxicity. Unfocused repetition creates false alarms, blood loss and workload without improving care.

The clinician receiving a result owns integration and action within local systems. Review critical communication, provisional status, comments and amended reports. If the result is incompatible with the patient, assess the patient first when harm is possible, then speak to laboratory staff about specimen and assay limitations. Pending cultures, histology and specialist tests must be listed at handover and discharge with a named reviewer and plan for communicating the result.

Key points

  • Confirm patient, sample time, specimen type, units, reference interval and whether the result is final, corrected, point-of-care or laboratory generated.
  • Ask why the test was ordered and what result was expected; an isolated flag is not a diagnosis and an unflagged result may still be dangerous.
  • Check pre-analytical threats such as wrong tube, haemolysis, contamination, delay, posture, fasting, infusion sampling and recent treatment before explaining physiology.
  • Interpret magnitude, pattern and trajectory together; related analytes and prior values usually discriminate artefact, adaptation and acute pathology better than one number.
  • Reference intervals describe a comparison population and assay, while decision limits answer particular clinical questions; neither replaces context.
  • Treat a surprising result as a verification problem: reassess the patient, compare priors, inspect flags, contact the laboratory and repeat only when it can change action.
  • Critical results and pending tests need closed-loop communication with documented receipt, action, patient information and ownership after transfer.
02Situations and prioritiesThe context, relevant information and actions that matter most.
Provenance

Identity, time, specimen, collection site, assay platform, units and report status determine whether values are comparable and attributable.

Pre-analytical discordance

Haemolysis, lipaemia, contamination, wrong container, delayed separation and sampling near an infusion can produce internally inconsistent results.

Magnitude and direction

Absolute deviation, rate of change and relation to the patient’s baseline often matter more than the red or green display flag.

Physiological pattern

Linked analytes should be interpreted as a system, using expected compensations and contradictions to test a proposed mechanism.

Reference versus decision limit

A population interval estimates expected distribution; a clinical threshold is tied to a validated decision and population.

Communication state

Identify critical calls, comments, provisional findings, corrected results and tests still processing, each with an action owner.

03Assessment and interpretationHow to gather information, assess the situation and recognise uncertainty.
Reasoning sequence

Consider the information, its meaning and its limitations before deciding what follows.

  1. 01
    Specimen validity review
    Why
    Check tube, collection site, timing, transport, haemolysis and laboratory comments.
    Interpretation and limitations
    A compromised specimen may require urgent clinical reassessment and a correctly collected repeat; it should not be silently averaged with valid results.
  2. 02
    Unit and reference verification
    Why
    Confirm units, assay-specific range and relevant age, sex or pregnancy context.
    Interpretation and limitations
    Never compare numbers across laboratories or eras until units and methods are compatible; unit error can be orders of magnitude.
  3. 03
    Pattern and delta analysis
    Why
    Compare linked analytes and previous values over a clinically plausible interval.
    Interpretation and limitations
    Concordant movement supports a physiological process; discordance should trigger scrutiny of sampling, timing, treatment and alternative mechanisms.
  4. 04
    Patient reassessment
    Why
    Repeat observations and focused examination when a result could imply immediate harm.
    Interpretation and limitations
    A critical biochemical result is an instruction to assess risk, not merely to order another specimen.
  5. 05
    Laboratory discussion
    Why
    Ask laboratory professionals about interference, uncertainty, specimen suitability and confirmatory methods.
    Interpretation and limitations
    Discussion can distinguish a repeatable analytical issue from genuine pathology and avoid an inappropriate downstream cascade.
  6. 06
    Purposeful repeat
    Why
    Repeat only with a defined timing, question and anticipated decision.
    Interpretation and limitations
    The interval should reflect disease tempo and test biology; minimum retesting guidance does not override urgent clinical need.
04Worked approachesCases with ordered reasoning, an action and a check of the outcome.
01Worked caseUnexpected severe hyperkalaemia flagA clinically stable patient has potassium 6.8 mmol/L on a haemolysed sample, unchanged creatinine and a previous potassium of 4.2 mmol/L.
  1. 1Recognise potential immediate harm, move the patient to a monitored setting, obtain an urgent 12-lead ECG and start continuous rhythm monitoring. A normal ECG or unchanged creatinine does not exclude dangerous genuine severe hyperkalaemia.
  2. 2Check sample source, haemolysis comment, collection difficulty, infusion proximity, timing and linked analytes; the discordant stable renal indices make contamination or haemolysis plausible but not proven.
  3. 3Escalate immediately, contact the laboratory and obtain a correctly collected urgent paired repeat; a rapidly available blood-gas potassium can guide action while formal laboratory confirmation is pending when locally validated and appropriate.
  4. 4If potassium is confirmed at 6.5 mmol/L or above, or severe hyperkalaemia is strongly suspected, start the current UK Kidney Association severe-hyperkalaemia pathway without waiting for ECG changes. If paired testing proves artefact, avoid unnecessary ongoing treatment; document the resolution and monitoring plan.
02Applied approachA small threshold crossingA surveillance result moves just beyond a treatment threshold after a recent change in assay provider.
  1. 1Verify units, method, reference or decision limit and whether the new laboratory is directly comparable with previous results.
  2. 2Estimate whether the change exceeds plausible analytical and within-person variation rather than treating the display flag as categorical truth.
  3. 3Review adherence, timing, intercurrent illness and medicines that could explain a real change.
  4. 4Repeat at a decision-relevant interval or act if the validated pathway says the threshold itself requires action, documenting uncertainty.
03Safety processPending result at dischargeA patient is clinically ready for discharge while cultures and a specialist antibody test remain outstanding.
  1. 1List each pending test, expected reporting time, clinical question and result that would change management.
  2. 2Assign a named clinician or service to review it and record how urgent findings will be communicated.
  3. 3Explain to the patient how and when results will be shared and what symptoms require earlier contact.
  4. 4Verify receipt and action after reporting; amend the care plan and communicate across settings when needed.
05Feedback, follow-up and evidenceReview outcomes, seek feedback and identify what to improve.
  • Trend only comparable results and annotate changes in method, specimen, treatment or timing that alter interpretation.
  • Review critical and corrected results until receipt and action are documented, including nights, weekends and discharge transitions.
  • Record why and when a repeat is needed, what result will trigger escalation and who will review it.
  • Monitor the patient rather than the analyte alone: symptoms, observations, organ function and treatment effects can diverge from one marker.
  • Review repeated low-value testing and blood loss, especially in stable inpatients with no new decision attached.
06Special situationsVariants, exceptions and circumstances that change the usual approach.

Flags are screening cues

Automated high and low labels compare with an interval; they do not encode urgency, causation, baseline or treatment threshold.

A repeat is not neutral

Repetition can delay treatment when danger is real and can amplify false positives when no decision or interval is specified.

Linked data test mechanisms

A proposed acid-base, endocrine, renal or hepatic mechanism should predict a coherent pattern; contradictions deserve explanation.

The laboratory is a clinical partner

Laboratory professionals can identify interference, sample problems, method changes and the best confirmation strategy.

Corrected reports matter

A later authorised or amended result can reverse an earlier plan; systems must surface and reconcile the change.

Examination reasoning

State the abnormal pattern, severity and trajectory, then offer a mechanism, key alternative, immediate risk and verification step.

Timing can explain discordance

Peak, trough, post-dose, fasting and post-procedure samples answer different questions; record timing relative to intervention.

Critical communication is bilateral

The laboratory transmits the result, while the clinical service must provide reachable contacts, acknowledge receipt and document action.

07Common pitfallsFrequent interpretation and management errors.
  1. 01

    Diagnosing from a single red flag without checking provenance, units or baseline.

  2. 02

    Dismissing a dangerous result as haemolysis without assessing the patient and confirming it.

  3. 03

    Comparing values from different methods or units as if the numerical change were biological.

  4. 04

    Repeating a test automatically without stating what decision the repeat will alter.

  5. 05

    Using a reference interval as though it were a validated treatment threshold.

  6. 06

    Discharging or handing over without ownership of pending, critical or corrected results.

Practice

Two practice questions

Question 1 of 20 correct
Clinical foundationsOriginal SBA

Critical discordant potassium

A stable inpatient has potassium 6.8 mmol/L on a sample flagged as haemolysed, with unchanged renal function and no symptoms. Which action best balances immediate safety and result verification?

Sources and review status4 sources · checked 7 Sept 2026 · clinical review pending
Sources

Sources and review status

National guidance is shown before implementation-dependent detail. Apply principles in context and verify current guidance when a decision affects care. Source check completed 7 Sept 2026; clinical approval remains outstanding.

Authoring stateComplete draftClinical stateAwaiting reviewJurisdictionUnited Kingdom