ISO 13528 Statistical Methods to be Used in Proficiency Testing Through Interlaboratory Comparison

The ISO 13528 standard, “Statistical methods for proficiency testing by interlaboratory comparison,” published by the International Organization for Standardization (ISO), details the statistical methods that proficiency testing providers can use to design proficiency testing schemes and analyze the data obtained from these schemes. This standard also provides recommendations for the interpretation of proficiency testing data by those involved in such proficiency testing schemes and by accreditation bodies.

The procedures in this standard can be applied by laboratories, inspection bodies, and individuals to demonstrate that measurement results obtained meet the specified criteria for acceptable performance. This standard applies to proficiency tests where the reported results are either quantitative measurements or qualitative observations relating to the test items.

The procedures in this standard may also be applied to the evaluation of expert opinions where opinions or judgments are reported in a manner that can be objectively compared with an independent reference value or a consensus statistic. For example, the provisions of this standard relating to nominal (qualitative) characteristics may be applied when classifying proficiency test items into known categories through inspection (or when determining through inspection whether proficiency test items originate from the same original source) and when the classification results are objectively compared.

For the purposes of this standard, proficiency testing refers to the assessment of participant performance against predetermined criteria through interlaboratory comparisons. In this standard, the term proficiency testing is taken in its broadest sense and includes, but is not limited to, the following:

Quantitative scheme. The aim is to determine the quantity of one or more magnitudes to be measured for each competency test item.
Qualitative schema: When the aim is to identify or define one or more qualitative characteristics of a proficiency test item.
Sequential scheme: Where one or more proficiency test items are distributed sequentially for testing or measurement and returned to the proficiency test provider at intervals.
Simultaneous scheme: In cases where proficiency test items are distributed for simultaneous testing or measurement within a specific time period.
One-time application: This is an application where all the elements of the proficiency test are provided only once.
Continuous scheme: In cases where proficiency test items are provided at regular intervals.
Sampling: Where samples are taken for later analysis and the purpose of the proficiency testing scheme involves evaluating the conduct of the sampling.
Data interpretation: Situations where data sets or other information are provided and the information is processed to provide an interpretation (or other conclusion).

In short, ISO 13528 is an international standard that provides guidelines for the use of statistical methods in proficiency testing programs conducted through interlaboratory comparisons. Subsequently revised significantly, this standard has become a fundamental reference for organizations that conduct proficiency testing or external quality assessment programs, and for the accreditation bodies that evaluate them (particularly within the scope of the ISO/IEC 17043 standard).

Proficiency testing is one of the most effective ways for laboratories to demonstrate the reliability of their measurement results. When hundreds of laboratories analyze the same test material and report their results, proficiency testing providers should be able to:

Assigning a reliable assigned value (the best estimate of the true value).
Calculating performance scores
To identify laboratories that are underperforming.
Doing all of this in a statistically sound and transparent way.

The ISO 13528 standard provides the statistical tools and rules necessary to do exactly that.

The 2022 revision of this standard introduced the following important updates:

Strong statistical performance: This is strongly highlighted by the new algorithms (algorithms A and S).
Processing multimodal data: A new section containing specific procedures has been added.
Assigned value from participants: Allows formulation only using specialized laboratories.
Uncertainty regarding assigned value: Mandatory evaluation and reporting have been introduced.
Performance scoring: Additional guidance has been added regarding combined scores and rescaled score totals.
Minimum number of participants: For robust methods, a minimum of 12-15 is recommended.

The basic statistical procedures in the ISO 13528 standard are as follows:

Determining the assigned value (X): Possible sources (in order of preference, where applicable):
Known value (certified reference material, formulation)
Reference values from one or more specialized laboratories.
Solid consensus from all participants.
Standard agreement (arithmetic mean) from participants
Standard deviation for proficiency assessment: Can be adjusted according to:
Regulatory or specification limits
Perception of suitability for purpose
Past proficiency test data
Solid standard deviation from the current tour.
Performance scores:
z-score
zeta score (includes measurement uncertainties)
The minimum score (a maximum of 1 is acceptable for test laboratories)
Robust methods: This standard now mandates the use of robust statistics for calculating consensus values and standard deviations when data is expected to contain outliers or come from slightly different populations.
Uncertainty of the assigned value: Proficiency test providers must now assess and report the standard uncertainty of the assigned value. If the assigned value is at least 0,3, the consensus value is considered highly uncertain, and alternative approaches (expert laboratories, reference materials) should be preferred.

The ISO 13528 standard is used by the following:

Proficiency test providers (to design and statistically evaluate their programs).
Accreditation bodies evaluate proficiency testing providers according to the ISO/IEC 17043 standard.
National metrology institutes.
Clinical, environmental, food, forensic, and calibration laboratories interpret proficiency test results.

In conclusion, the ISO 13528 standard represents the current international consensus on how laboratory performance should be fairly and robustly assessed in inter-laboratory comparisons. Its increased emphasis on robust statistics, mandatory uncertainty assessment, and clearer rules for multimodal data makes proficiency testing results more reliable than ever before. Familiarity with this standard is crucial for any organization running or participating in proficiency testing programs.

For many years, our organization, with its competent and experienced staff, has been understanding the needs of its clients, providing the advanced testing services they require, and assisting them in the establishment, implementation, and improvement of management systems. Within this framework, we also provide testing services to businesses in accordance with the "ISO 13528 Statistical methods for proficiency testing through interlaboratory comparison" standard.