How to Implement ISO 15189 in a Medical Laboratory
Build a diagnostic laboratory system around the patient the result belongs to.
What a how-to is not
A how-to is SPEQ’s practitioner method, not a procedure. It does not replace your own SOP, it is not a validated approach, and the judgement calls in it belong to your quality unit.
The medical laboratory standard governs testing performed on human specimens for the care of an individual patient. Its distinguishing feature is that the requirements extend beyond the analysis into the pre-examination and post-examination phases, where most diagnostic error actually happens, and that risk is assessed in terms of harm to the patient rather than defects in the result. Its current edition also brought point-of-care testing inside its scope.
- 1
Scope the system across all three examination phases
Requesting, patient preparation, collection, transport and receipt; examination; then reporting, interpretation and result communication. Building a system that starts at the bench leaves the phases responsible for most error outside the quality system.
- 2
Assess risk in terms of harm to the patient
The question is what happens to a patient if this goes wrong, not how often a process fails. That framing changes priorities — a rare error on a result that drives an irreversible clinical decision outranks a frequent one on a result that is routinely repeated.
- 3
Bring point-of-care testing inside the system
Testing performed near the patient by staff who are not laboratory personnel is within scope in the current edition, which absorbed the separate point-of-care standard. That usually means training, competence and quality control for people who do not report to the laboratory, which is an organisational problem before it is a technical one.
- 4
Establish competence per examination and reassess it
Competence is evidenced for the specific examinations a person performs, with reassessment at defined intervals. A general induction record does not evidence competence for a particular examination on a particular day.
- 5
Verify or validate every examination method
Verify that a validated commercial method performs as claimed in your hands, with your instruments, staff and population; validate methods you develop or modify. A method verified in the manufacturer’s population is not verified in yours.
- 6
Control internal quality control, external assessment and turnaround
Internal control detects drift, external assessment detects bias against peers, and turnaround is a clinical requirement rather than a service metric — a correct result that arrives after the decision was made has not served the patient.
- !A quality system that starts at the bench, leaving the phases where most diagnostic error occurs outside it.
- !Point-of-care testing treated as outside the laboratory, when the current edition brought it inside.
- !A commercial method accepted on the manufacturer’s validation without verification in your own population.
- !Turnaround managed as a service metric rather than as a clinical requirement.
How to Implement ISO 15189 in a Medical Laboratory: frequently asked questions
Common questions on implement iso 15189 in a medical laboratory.
How does this differ from the testing-competence standard?
That standard addresses competence in testing and calibration generally. This one is written for medical laboratories: it extends into the pre- and post-examination phases, frames risk as harm to the patient, and treats the clinical usefulness of the result as part of the requirement. The current edition is aligned in format, which makes the difference in purpose easier to overlook.
Does it apply to point-of-care testing?
Yes. The current edition absorbed the separate point-of-care standard, so testing performed near the patient is within scope. In practice this is an organisational challenge, because the people performing it usually do not report to the laboratory that is now accountable for it.
Can a diagnostic accreditation cover clinical trial samples?
No. It covers testing for the care of an individual patient. Analysing samples to generate research data against a trial protocol is a different activity with different obligations, and a laboratory doing both carries both — which is a design decision, not something to sort out when asked.