The language of QC, EQA and accreditationEQC guide · Glossary
Short definitions of the terms used in internal quality control, external quality
assessment and accreditation, in alphabetical order within each group.
Quality control
Assayed control
Control material supplied with target values and ranges per method. Unassayed controls come without them and need targets set by the laboratory.
Control limits
The range, usually the mean ± 2 or 3 SD, inside which a control result is accepted. Best calculated from the laboratory’s own data.
Error detection
The probability that the QC rules reject a run that contains an error of a given size.
False rejection
A run rejected by the QC rules although the method works correctly. A 1-2s rule produces about one false rejection in twenty measurements.
Internal quality control (IQC)
Measuring control material together with patient samples to decide whether a run can be released. See Internal quality control.
Levey-Jennings chart
A chart of control results over time with the mean and SD limits as lines; the standard view for spotting shifts and drifts.
Multirule QC (Westgard rules)
A combination of control rules such as 1-3s, 2-2s, R-4s, 4-1s and 10x, each sensitive to a different kind of error, with few false rejections.
Run
The interval, defined by time or number of samples, for which one QC evaluation decides whether results are released.
Third-party control
Control material from a manufacturer independent of the analyser and reagent supplier; it shows calibration shifts that matched controls can follow.
External quality assessment
Accuracy-based survey
An EQA scheme whose targets come from reference methods and commutable samples, so it assesses trueness and not only agreement with peers.
Commutability
The property of a control or EQA sample to behave like a patient sample across different methods. Non-commutable material can make a correct method look biased.
External quality assessment (EQA)
Analysing samples of unknown value from an outside provider and comparing the results with targets and other laboratories. Mandatory under ISO 15189 for every examination.
Interlaboratory comparison
Any organised comparison of results between laboratories; EQA is one form, sample exchange between two laboratories another.
Peer group
The laboratories in an EQA scheme that use the same method, analyser and often reagent, and against whose results a laboratory is compared.
Proficiency testing (PT)
The term used in the United States and in accreditation standards for EQA with graded results; in practice used interchangeably with EQA.
Ringversuch
The German term for an EQA distribution, used for example by INSTAND and in the RiliBÄK guideline.
Statistics and performance
Bias
The systematic difference between the laboratory’s results and the true or target value, usually in percent. EQA is the main source for estimating it.
Coefficient of variation (CV)
The standard deviation divided by the mean, in percent; the usual measure of imprecision.
CV ratio (CVR)
The laboratory’s CV divided by the CV of its peer group; above 1.5 means noticeably higher imprecision than comparable laboratories.
Measurement uncertainty
A parameter describing the dispersion of the values that can reasonably be attributed to the measured quantity, usually given as an interval around the result. ISO 15189 requires it for quantitative examinations.
Sigma metric
(TEa − |bias|) / CV, all in percent. It expresses how many SDs of the method fit into the allowable error and guides the choice of QC rules.
Standard deviation index (SDI)
The difference between a laboratory’s result and the peer-group mean, divided by the peer-group SD. Values beyond ±2 call for an investigation.
Total allowable error (TEa)
The largest error a result may have without affecting its clinical use; taken from regulations such as CLIA, from biological variation or from national rules.
z-score
The difference between a result and the assigned value of an EQA sample in units of the SD for proficiency assessment; |z| ≤ 2 is satisfactory, |z| ≥ 3 unsatisfactory.
Standards and organisations
CAPA
Corrective and preventive action: the documented process of finding the cause of a nonconformity, removing it and checking that the action worked.
CLIA
The Clinical Laboratory Improvement Amendments, the US federal rules for clinical laboratories, including acceptance limits for proficiency testing.
College of American Pathologists (CAP)
A US organisation that runs proficiency testing programmes and a laboratory accreditation programme used worldwide. KombiMED is its official distributor for Central Asia and the Caucasus. See CAP accreditation.
ISO 15189
The international standard for quality and competence of medical laboratories; current edition 2022. See ISO 15189 and accreditation.
ISO/IEC 17025
The standard for testing and calibration laboratories, used for food, environmental and pharmaceutical testing.
ISO/IEC 17043
The standard for the competence of proficiency testing providers; accreditation to it is a sign of a reliable EQA scheme.
Point-of-care testing (POCT)
Testing near the patient, outside the central laboratory. Since ISO 15189:2022 it falls under the laboratory’s quality management, including EQA.
RiliBÄK
The guideline of the German Medical Association on quality assurance in medical laboratories, with rules for internal QC and EQA participation.
