Some of the new provisions of ISO/IEC 17025 are similar to those of ISO 15189. These refer to risks and opportunities (risk-based thinking vs risk management), control of data and information management (more detailed requirements, addressing new technology tools), the competence of proficiency testing schemes of providers and reference materials producers based on their compliance with the relevant standards, i.e. ISO/IEC 17043 (2010) and ISO 17034 (2016) respectively.
Although the revision of ISO 15189 is still going on, some comments on the possible outcome can be made. It is considered that some of the new provisions of ISO/IEC 17025 will be included in the revised edition of the standard for medical laboratories, adapted as appropriate.
Uncertainty from Sampling in ISO/IEC 17025
The inclusion of sampling as a stand-alone laboratory activity, although not expressed in this way, represents one of the main changes compared with the 2005 version of ISO/IEC 17025. That is reflected in several other provisions of the new standard, mainly those referring to measurement uncertainty (Tsimillis, 2018).
Sub-clause 7.6.1 specifies that, when evaluating measurement uncertainty, all contributions of significance, including those arising from sampling, shall be considered.
Sub-clause 7.8 specifies the requirements for reporting results; sub-clause 7.8.2 includes standard requirements for all reports, namely test, calibration or sampling. Among others, it is required that the date of sampling shall be included where this is critical to the validity and application of the results (sub-clause 7.8.2.1.h); similarly, reference to the sampling plan and sampling method shall be included where these are relevant to the validity or application of the results (sub-clause 7.8.2.1.k). If the laboratory has not been responsible for the sampling stage, a statement shall be included that the results apply to the sample as received (sub-clause 7.8.2.2). Additional requirements for each type of the said laboratory activities are included in sub-clauses 7.8.3 (testing), 7.8.4 (calibration) and 7.8.5 (sampling). The latter applies if the laboratory is responsible for the sampling activity and, where necessary, for interpreting results. Points (a)-(e) refer to details of the sampling procedure that shall be included in the report, i.e. date, identification of the item or material sampled, the location of sampling, the sampling plan and method, as well as details of any environmental conditions during sampling that affect the interpretation of the results. Last but not least, according to point (f), information required evaluating measurement uncertainty for subsequent testing or calibration shall be included in the report as well.
Testing laboratories trying to meet ISO/IEC 17025 requirements may find some approaches addressing the uncertainty from sampling (Ramsey et al., 2019; Magnusson et al., 2020) helpful. These approaches are applicable when the measurand is defined in terms of the analyte concentration in the sampling target, rather than just the sample delivered to or tested by the laboratory. These approaches are based on replicate sampling, which is appropriate for analytical laboratories, but they may not readily apply in all cases in medical laboratories, as we shall see later. At the same time, trying to answer this presentation's central question, we need to elaborate on the requirements of ISO 15189 for sampling and measurement uncertainty.