ISO 15189:2022 · Section 7 · Process requirements
Clause 7.3.4: Evaluation of measurement uncertainty
Check the exact requirement in your licensed copy of the standard.
Last reviewed
What this clause is about
Measurement uncertainty describes the dispersion reasonably associated with a measured quantity under defined conditions. It helps the laboratory understand how much variation may accompany a result and whether the method is suitable for its clinical purpose. It is not a correction added to every result, a tolerance for poor performance, or a guarantee that the patient's true value lies within a particular interval.
Identify the measured quantities for which uncertainty evaluation is relevant and define the scope of each estimate. Consider the examination method, concentration range, specimen context, and reporting units. Where evaluation is not meaningful or possible, document the reasoning rather than leaving an unexplained gap. For examinations producing non-numerical outcomes, consider the uncertainty of relevant numerical steps when these influence classification and require evaluation.
Use an approach suited to the method and available evidence. Long-term internal quality control data can provide information about routine imprecision, while calibration information, reference materials, or other studies may identify additional significant contributions. Avoid both omitting an important component and counting the same source twice. Explain assumptions, data selection, coverage factor where used, and how the resulting estimate applies across concentrations or decision points.
Compare the estimate with requirements relevant to the intended clinical use and review it when performance or significant inputs change. Ensure staff can explain its meaning to users and make information available when appropriate or requested. LabToolkit Pro can organize uncertainty worksheets, supporting records, and review approvals; the laboratory must choose a technically justified approach and interpret its clinical implications.
What assessors typically ask to see
- Which measured quantities have an uncertainty evaluation, and how is the scope or any justified exclusion documented?
- What data represent routine performance over relevant operators, reagent lots, calibration events, and other normal sources of variation?
- How were significant uncertainty contributions identified, and what prevents omissions or double counting within the chosen calculation approach?
- How are concentration dependence, measurement units, coverage assumptions, and limitations described so users do not misinterpret the estimate?
- How is the estimate assessed against requirements relevant to clinical use, particularly near important medical decision points?
- What triggers review, and who can explain the evaluation and its limitations when a clinical user requests information?
Documents and records that usually provide evidence
- Uncertainty procedure describing scope, responsibilities, selected evaluation approaches, required inputs, review triggers, and communication arrangements for users.
- Source data records identifying control materials, observation periods, calibration information, relevant reference values, and reasons for data selection.
- Calculation worksheets showing units, components, assumptions, combination steps, coverage factor where applicable, and the resulting uncertainty estimate.
- Technical assessment relating the estimate to intended use, performance requirements, concentration dependence, and important limitations in interpretation.
- Authorized review records confirming data suitability, calculation checks, decisions about acceptability, and actions where further investigation is needed.
- Change and communication records showing updates after significant performance changes and responses to requests for uncertainty information.
Ready-made forms mapped to clause 7.3.4: Measurement Uncertainty Reporting Form, Measurement uncertainty calculator.
Common findings to avoid
- A calculation is retained without stating which examination, concentration range, units, or operating conditions it actually represents.
- Short control runs are treated as long-term performance evidence despite missing routine calibration changes and reagent lot variation.
- The same source of variation appears in several components, producing an inflated estimate without an explanation of overlap.
- Uncertainty is calculated once and never reconsidered after persistent performance changes or a substantial calibration modification.
How LabToolkit Pro maps documents to this clause
Each SOP, form and record in LabToolkit Pro can carry clause tags. When a document tagged 7.3.4 becomes effective, it counts as direct evidence on the readiness dashboard for ISO 15189:2022; links from other standards through the crosswalk are shown separately as indicative coverage. Gaps appear clause by clause, so you can see which parts of 7.3.4 still need a controlled document or a current record. Readiness is a planning aid — only an accreditation body confirms conformity.
Frequently asked questions
Is measurement uncertainty the same as allowable error?
No. An uncertainty estimate describes dispersion associated with measurement under specified conditions. An allowable error limit is a requirement or acceptance boundary selected for a particular purpose. They may inform a performance assessment together, but they answer different questions and should not be substituted for one another in records or explanations.
Must uncertainty appear on every patient report?
Not necessarily. Establish how uncertainty information is made available and when reporting it is relevant to interpretation or specifically required. The laboratory should be able to explain the estimate when requested. Avoid adding unexplained numbers to reports where they could create confusion rather than support a clinical decision.
Can internal quality control data be used?
Yes, when the material, observation period, and performance conditions provide suitable information for the chosen approach. Consider what the data capture and what they miss, including relevant systematic contributions. Explain any assumptions about applicability to patient specimens and concentrations, and revise the evaluation when the underlying performance information changes.
Check the exact requirement in your licensed copy of the standard.
Put it into practice
Draft, approve and e-sign your SOPs with version control, or get the complete template pack.