33. The Committee recalled that this Agenda Item had been proposed at the last Session due to a concern that a number of international organizations and accreditation agencies were developing recommendations and requirements regarding measurement uncertainty which were at variance with present practices of Codex[24]. The Committee had requested the Executive Committee to approve the initiation of work of this issue, which had been subsequently granted.
34. In the presentation of the paper, the Delegation of the United Kingdom introduced the problems encountered in the country in relation to the accreditation of laboratories which required an estimation of measurement uncertainty, and the possibility that the use of the term "measurement uncertainty" could be misleading. The Delegation introduced an ongoing project within the UK Ministry of Agriculture, Fisheries and Food to assess the ISO measurement uncertainty approach[25] which was complicated and time-consuming. The Committee was informed that EURACHEM was attempting to utilize the measurement uncertainty approach and it would review the ISO Guide[26] in July of this year.
35. Several delegations indicated that the ISO Guide was heavily oriented toward metrology and, as a result, was either too stringent in requirements or not suitable for application in food analysis. Several delegations expressed the opinion that food analysis laboratories accredited to ISO Guide 25 should be allowed to utilize method-performance data or, if it was not available, internal quality control or method validation data to estimate measurement uncertainty.
36. The Committee therefore agreed to the following:
1. The Committee will develop for Codex purposes an appropriate alternative term for measurement uncertainty, e.g. measurement reliability.37. The Committee agreed to request the Delegation of the United Kingdom to redraft the paper for consideration by the Committee at its next Session. The Delegation of the United Kingdom invited contributions of other delegations.2. The precision of a method may be estimated through a method-performance study, or where this information is not available, through the use of internal quality control and method validation data.
3. Consideration should be given as to whether it is necessary to undertake an additional formal evaluation of a method of analysis using the ISO approach in addition to using information attained through a collaborative trial.
4. Governments should advise accreditation agencies that for national and Codex purposes the measurement uncertainty of a result need not be calculated using the ISO approach provided the laboratory is complying with the appropriate Codex principles.
38. The Committee was informed that ISO Guide 25 was under revision and was currently receiving public comments. It was also informed of a forthcoming meeting of International Laboratory Accreditation Conference (ILAC) in Paris in April 1997, and that the Committee's interests would be conveyed at that meeting.
