Templates and records · ISO 15189 · Clause 7.3.7.2

ISO 15189 medical laboratory internal quality control

Q: Does ISO 15189 require Westgard rules?

No. It requires defined criteria for accepting or rejecting runs and detecting trends. Westgard multirules are the most widespread, but the laboratory can design its own based on the method's performance and the error it needs to detect.

Q: What changes compared with quality control under ISO/IEC 17025?

The logic is the same, but ISO 15189 adds that levels should be close to clinical decision limits, that frequency should depend on the risk to the patient and that, after a failure, patients already tested should be reviewed.

The controls run every day on every analyser, chosen to detect in time an error that could change a clinical decision, and what is done about patients when a control fails.

In one sentence

A medical laboratory's internal quality control record captures control materials at concentrations close to clinical decision limits, their frequency according to the risk to the patient, the acceptance rules, each result with its trend and, after a failure, the decision not to release results and the review of patients tested since the last acceptable control, because clause 7.3.7.2 of ISO 15189:2022 requires that procedure and the detection of trends.

Content reviewed on 17 October 2026

Where the standard requires it: ISO 15189 · Clause 7.3.7.2 Clauses are cited using the numbering of ISO 15189:2022, the edition in force since the transition ended in December 2025 and the same one the Academy uses.
Type of information: Mandatory record Internal QC data reviewed against defined criteria (7.3.7.2)
Who fills it in: Analyser technician The section consultant reviews trends and decides on patients after a failure
When: With each run or at the defined frequency At least at the start of each shift on continuous analysers; monthly trend review
How long it is kept: Set by the laboratory ISO 15189 asks for the retention time of each record to be defined, taking into account legal and user requirements (8.4). Those forming part of the medical record also follow the applicable health regulations.

What the standard requires

What ISO 15189 demands and what it leaves to your judgement

The standard requires an internal quality control procedure using materials that behave as closely as possible to patient samples, at concentrations at or near clinical decision limits, and at a frequency based on the stability of the method and the risk of harm to the patient. It recommends considering independent third-party materials.

Data are reviewed against defined criteria to detect both random errors and trends. When the criteria are not met, the laboratory does not release the affected results and evaluates those of patients tested since the last acceptable control.

That retrospective review of patients is the big practical difference from ISO/IEC 17025, where control demonstrates the validity of the method; here, it also protects specific people.

Read it in context: Ensuring the validity of results, in the ISO 15189 Academy

Structure

Fields the template should include

Measurand, analyser and reagent
Control material, lot and levels
Laboratory's own mean and SD for each lot
Control rules applied
Results by date and time
Control failure: rule violated and cause
Decision on the run and results held
Review of patients since the last acceptable control
Monthly trend review and cumulative coefficient of variation

Example

Example: glucose control failure

Company, people and figures are fictitious: the example only shows what a properly completed record looks like.

Field	Content
Measurand	Serum glucose · analyser 2 at Riverside clinical laboratory (fictitious)
Control	Independent third-party material · levels of 90 and 250 mg/dL
Rules	1-3s reject, 2-2s reject, 1-2s warning
Failure	15 May, 16:00 · high level at +3.2 SD · rule 1-3s
Cause	Bubble in the reagent probe
Patients	86 since the acceptable control at 12:00 · 20 random samples retested: differences < 3%
Decision	Results issued are upheld · section consultant
May review	Cumulative coefficient of variation 1.8% · no trends

At the audit

Mistakes auditors commonly flag

Controls repeated until they "pass" without investigating the failure.
No review of patients tested since the last acceptable control.
Control levels far from the clinical decision limits.
The manufacturer's mean and SD used instead of the laboratory's own.
Only materials from the reagent manufacturer, which can mask drifts.

In Kimobox

How this record is kept in Kimobox

1Internal QC results are recorded with their trend, visible on a dashboard so a drift can be caught in time.
2A control failure is recorded as nonconforming work, with its assessment of the impact on previous results.
3Clinical reports are generated from a controlled template.

Frequently asked questions

What people ask us before getting started

Does ISO 15189 require Westgard rules?

No. It requires defined criteria for accepting or rejecting runs and detecting trends. Westgard multirules are the most widespread, but the laboratory can design its own based on the method's performance and the error it needs to detect.

What changes compared with quality control under ISO/IEC 17025?

The logic is the same, but ISO 15189 adds that levels should be close to clinical decision limits, that frequency should depend on the risk to the patient and that, after a failure, patients already tested should be reviewed.

Keep your ISO 15189 records without loose templates

We'll show you the module running with data similar to yours.

Request a demo

Keep exploring