Originally Published IVD Technology January/February 2006
REGULATIONS & STANDARDS
Recent progress in metrological traceability
Richard Miller
 |
| Richard Miller is a staff scientist in consumables manufacturing at Dade Behring Inc. (Deerfield, IL). He can be reached at rick_miller@dadebehring.com. |
Since the implementation of the European Union’s Directive on In Vitro Diagnostic Medical Devices (IVD Directive), the traceability of IVDs has come under increased focus. IVD manufacturers are now required to make their calibrators traceable to reference materials or methods of a higher order, where they are available, if they intend to sell their products in the EU.
There has also been a focus on the traceability of the laboratory test results themselves. In fact, clinical laboratories in many countries are required to provide information about the traceability of their test results, which has some impact on IVD manufacturers as well.
To provide guidance in meeting the IVD Directive’s requirements, the European Committee for Standardization (CEN; Brussels) and the International Organization for Standardization (ISO; Geneva) prepared several standards. Approved in 2002 and 2003, these standards describe the requirements for demonstrating traceability to higher-order reference materials or methods.1–5 If IVD manufacturers comply with these standards, they are presumed to be in compliance with the IVD Directive.
Where reference materials or methods of a higher order exist, IVD manufacturers are required to apply them in the value assignment of their calibrators. As a result, manufacturers may have to revise the design of their calibrators and controls to include such materials or methods in their value assignment systems. IVD manufacturers must also demonstrate the commutability of their calibrators and establish the uncertainty of the values assigned to them. In addition, manufacturers must report such information to customers upon request. Further information on traceability can be found in various articles that have been previously published in IVD Technology.6,7
Several efforts are under way to develop documents that will assist IVD manufacturers in complying with the traceability requirements. Such efforts have focused on the following two areas:
• Identifying reference materials, reference methods or procedures, and the reference laboratories that perform them.
• Developing guidance documents that describe how to meet the ISO/ CEN standards.
Identifying Reference Materials and Methods
To demonstrate traceability to higher-order reference materials or methods, IVD manufacturers need to know what materials and methods are available. The IVD Directive has recognized that appropriate materials or methods may not be available for every measurand or analyte for which manufacturers make calibrators and controls.
To assist IVD manufacturers in determining what methods and materials are available, the International Bureau of Weights and Measures (BIPM; Paris), the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC; Milan, Italy), and the International Laboratory Accreditation Cooperation (ILAC; Rhodes, Australia) established the Joint Committee for Traceability in Laboratory Medicine (JCTLM). JCTLM’s mission is to “support worldwide comparability, reliability, and equivalence of measurement results in laboratory medicine.” JCTLM accomplishes this mission by identifying reference materials, reference procedures, and reference laboratories that meet the ISO standards mentioned above.
JCTLM formed two working groups to conduct the identification of the reference materials, methods, and labs. Working group 1 is responsible for identifying the reference materials and methods. This working group uses ISO 15193 and 15194 as the criteria for determining which materials and methods are of a higher order.3,4 Working group 2 is responsible for identifying the reference laboratories, and uses ISO 15195 as its criterion.5 Both working groups use 13 review teams that consist of international experts in specific analyte categories (e.g., metabolites, enzymes, electrolytes, nucleic acids). Each review team also includes representatives from IVD manufacturers, the national metrology institutes, accreditation organizations, and professional societies from the United States, Europe, and the Asia-Pacific region.
Working group 1 solicited a first round of nominations for higher-order reference materials and procedures, and reviewed them. As a result of this first round, the working group generated List I, which includes those materials and procedures that meet the traceability requirements. So far, working group 1 has listed reference materials for 106 analytes that are traceable to the international system of units (SI), and reference methods for 48 SI-traceable analytes.
Working group 1 also prepared List II, which includes those reference materials that are not SI traceable. Such materials either have been assigned values by an internationally agreed protocol or are purified substances in which the commutability cannot be determined because there are no reference procedures for the analytes. This list identifies materials for 18 analytes. The preamble to List I of higher-order reference materials and measurement procedures should be read before using the materials in Lists I and II.
As one of its operating principles, JCTLM set forth the need to develop a process for selecting reference materials, methods, and laboratories that is fair, technically valid, and able to be understood by all interested parties. With such goals in mind, working group 1 prepared a quality manual that carefully describes the process for soliciting, reviewing, and listing nominated reference materials and procedures. Where more than one material is available, the manual describes a process for evaluating the comparability of those materials. This process could be considered a final check on the validity of the nomination and listing processes. To date, materials for two analytes, cholesterol and potassium, have gone through this process.
Working group 1 has solicited and received a second round of nominations for reference materials and methods. The working group reviewed these nominations, and posted those that meet the traceability requirements at the end of last year. Working group 1 is constantly developing lists of materials and methods of a higher order. Since such lists are changing approximately every year, IVD manufacturers must review them periodically to be aware of additions.
Identifying Reference Laboratories
Working group 2 has not progressed as far in its tasks as working group 1. One reason is that the identification of the reference laboratories by working group 2 depends on the identification of the reference methods by working group 1. Since List I became available, working group 2 has been completing its quality manual. Working group 2 has also solicited and received 337 nominations for candidate calibration laboratories to measure 93 analytes. Since working group 1 has listed reference methods for only 48 analytes, not all of the nominated laboratories are using reference methods that meet the traceability requirements. Once the quality manual is completed, working group 2 will start reviewing each of the nominated laboratories according to the criteria in ISO 15195.
Another issue that working group 2 has been discussing is assuring compliance of the appropriate reference laboratories with ISO 15195. Working group 1 developed a system of assuring compliance of the reference materials with ISO 15194 that could be carried out by the review teams. Meanwhile, working group 2 stated that its goal is to list only the reference measurement laboratories that have accreditation either according to ISO 15195 or as calibration laboratories under ISO 17025. Working group 2 has also set a goal of having all laboratories accredited within three years.
Of the 337 nominated laboratories, only five are accredited as calibration laboratories according to ISO 17025. Some of these laboratories are part of existing networks that do not require accreditation to ISO 15195 or 17025. Such networks may even use different criteria to accept laboratories as members of the network. Since some laboratories may perform only one or two reference procedures, putting in the effort to become accredited will add significant costs to providing their services, which would be passed on to their customers. Although accreditation is an acceptable mechanism for identifying laboratories that do not meet the proper ISO traceability requirements, it may eliminate those laboratories that are already providing acceptable services but cannot afford the cost of accreditation.
To address the issue of credentials for calibration laboratories, the National Institute of Standards and Technology (NIST; Gaithersburg, MD) and the industry division of the American Association for Clinical Chemistry (AACC; Washington, DC) held a discussion forum at AACC’s annual meeting last July. This forum was intended to identify other options for demonstrating compliance to ISO 15195 that will be more cost-effective to reference measurement laboratories, and to propose other solutions to working group 2. The result of this forum was a proposal to the JCTLM executive committee to allow the development of an accreditation program for reference laboratory networks. According to this proposal, the leaders of the networks will be responsible for assuring compliance of their individual laboratories with the appropriate ISO standards. Since most of these laboratories are associated with networks, this proposal has the potential of reducing the costs for each laboratory while still meeting the goal of compliance.
Other Traceability Activities
IFCC. One requirement for calibration laboratories under ISO 15195 is participation in periodic interlaboratory comparisons. IFCC formed JCTLM to assist in developing reference laboratory networks. The committee has performed two trial comparison exercises and is soliciting participation in a third trial. The results of the first trial have been reported in a JCTLM document, “IFCC Ring Trial Results.”
CLSI. While ISO 17511 and 18153 describe general principles for establishing, validating, and documenting traceability, they do not provide detailed guidance.1,2 To address this problem, the Clinical Laboratory Standards Institute (CLSI; Wayne, PA) and IFCC have partnered to produce a guidance document for IVD manufacturers. CLSI and IFCC are developing a document, “Metrological Traceability and Its Implementation: A Report,” that is currently in draft form and is being revised to address comments solicited and received from experts.
CLSI has also initiated another project in the area of qualification of reference materials. Several ISO guidelines describe the requirements for reference material producers regarding quality systems, value assignment, and the preparation of reference materials. However, while such documents are available for all types of reference materials, whether physical or chemical in nature, very few address the specific needs of the clinical measurement community, such as the commutability of reference materials. For example, it is not clear when national metrology institutes need to demonstrate the commutability of a material and what the approaches for doing so are. If an IVD manufacturer or a clinical laboratory uses a reference material that is not commutable, it will not be able to provide traceable results; also, commutability data often are not available from the reference material producers. To address this problem, CLSI has initiated a publication project, “Validating and Implementing Secondary Reference Materials,” that will focus on such issues related to reference materials. This document will be helpful to IVD manufacturers, national metrology institutes, and some clinical and reference laboratories in determining what additional validation data may be required.
ISO TC 212. Originally approved in 2002, several of the ISO/CEN traceability standards will need to be revised in the near future. Consequently, ISO TC 212’s working group 2 is being reactivated to review and revise the relevant standards. Since ISO and CEN have a mutual recognition agreement that allows collaboration on such international standards, the revisions will be done in conjunction with CEN TC 140. To date, ISO 15193 and 15194 need revision, and activity on ISO 15195 will be initiated in 2006.3–5 IVD manufacturers that are interested in commenting on these standards should contact their country’s national standards organization. In the United States, the contact organization is CLSI.
Traceability for Clinical Laboratories
ISO TC 212 also approved ISO 15189, “Medical Laboratories—Particular Requirements for Quality and Competence,” which describes the quality system requirements for clinical laboratories. This standard requires clinical laboratories to have traceability programs for their results and to maintain programs for calibrating and determining the uncertainty of results where appropriate. In some countries, laboratories are being audited for compliance with these standards. As a result, some laboratories are asking IVD manufacturers for their traceability and uncertainty information, and for guidance in meeting the standards.
Although IVD manufacturers are not required to provide such guidance, it is their responsibility to provide the laboratories the traceability information for the values assigned to their calibrators. Where appropriate, such information includes the uncertainty of the values assigned to the calibrators.
Clinical laboratories are then faced with documenting the traceability and uncertainty of their test results. Although much of the traceability information will come from IVD manufacturers, the laboratories themselves must determine the uncertainty of the results obtained on patient samples.
To help with this issue, CLSI and ISO TC 212 initiated the following two activities. CLSI formed a subcommittee to prepare a guidance document on how to determine the uncertainty of measurement results, “The Expression of Uncertainty of Measurement.” Recruited and organized in the fall of 2004, this subcommittee produced its first draft last fall. ISO TC 212’s working group 2 also initiated a project, “Laboratory Medicine—Estimation and Reporting of Measurement Uncertainty of Routine Values,” which is described as an international standard outlining the principles of estimating measurement uncertainty. At the ISO TC 212 meeting last May, both the chair of the CLSI project on uncertainty and the convener of ISO TC 212’s working group 2 agreed to collaborate to ensure that these two documents are in alignment.
These two efforts will increase the laboratories’ interest in obtaining traceability information about the calibrators and controls they get from IVD manufacturers. In order to assess accurately the uncertainty of test results, laboratories will also need to obtain information that manufacturers may not have provided in the past. For example, such information might include the comparison of their reference methods with a procedure of a higher order, uncertainty information, and the like.
Conclusion
Substantial progress has been made toward supporting IVD manufacturers in meeting the IVD Directive’s requirements for establishing, validating, and documenting the traceability of their calibrators and controls. The work of JCTLM, IFCC, and CLSI should be recognized as a major help to manufacturers. However, despite the implementation of these requirements in 2003, more progress still needs to be made. Documents providing further guidance on complying with the directive are being developed and prepared. Such documents will assist IVD manufacturers in meeting these requirements in a uniform manner.
With the implementation of ISO 15189, clinical laboratories in some countries are being asked to provide the same type of traceability information to not only their auditors but also their end-users. This requirement has increased the clinical laboratories’ interest in obtaining traceability information from IVD manufacturers, which will continue. There is also a question as to how clinical laboratories and physicians will use such information. This issue is open to significant interpretation, and further guidance is needed.
As the guidance documents discussed in this article are prepared for IVD manufacturers and the clinical laboratories, efforts must be made to ensure that they are consistent with the appropriate ISO documents. Such consistency in the guidance documents is required so that inconsistencies and conflicts are not introduced into the system.
References
1. In Vitro Diagnostic Medical Devices—Measurement of Quantities in Samples of Biological Origin—Metrological Traceability of Values Assigned to Calibrators and Control Materials, ISO FDIS 17511 (Geneva: International Organization for Standardization, 2002).
2. In Vitro Diagnostic Medical Devices—Measurement of Quantities in Biological Samples—Metrological Traceability of Values for Catalytic Concentration of Enzymes Assigned to Calibrators and Control Materials, ISO FDIS 18153 (Geneva: International Organization for Standardization, 2002).
3. In Vitro Diagnostics Systems—Measurement of Quantities in Samples of Biological Origin– Preparation of Reference Measurement Procedures, ISO FDIS 15193 (Geneva: International Organization for Standardization, 2002).
4. In Vitro Diagnostic Medical Devices—Measurement of Quantities in Samples of Biological Origin—Description of Reference Materials, ISO FDIS 15194 (Geneva: International Organization for Standardization, 2002).
5. In Vitro Diagnostic Medical Devices—Measurement of Quantities in Samples of Biological Origin—Requirements for Reference Measurement Laboratories in Laboratory Medicine, ISO FDIS 15195 (Geneva: International Organization for Standardization, 2002).
6. DM Powers, “Coming Together on Traceability,” IVD Technology 8, no. 6 (2002): 22–30.
7. R Miller, “Metrological Traceability Requirements under the IVD Directive,” IVD Technology 9, no. 7 (2003): 23–31.
Copyright ©2006 IVD Technology
