Review

The Joint Committee for Traceability in Laboratory Medicine (JCTLM):

A Global Approach to Promote the Standardisation of Clinical
Laboratory Test Results
*David Armbruster,,1 Richard R Miller2
1
Global Scientific Affairs, Abbott Diagnostics, Abbott Park, IL 60064, 2Consumables Manufacturing, Dade Behring,
Glasgow, DE 19711
*For correspondence: David Armbruster e-mail: [email protected]

Abstract
Clinical laboratories are moving towards global standardisation to produce equivalent test results across space and time.
Standardisation allows use of evidence-based medicine, eliminates the need of method-specific reference intervals, decision
levels and cut-offs, and can be achieved by application of metrological principles. For example, in vitro diagnostics (IVD)
manufacturers can make kit calibrators traceable to internationally recognised reference materials and reference methods.

The first step towards standardisation is to identify appropriate reference materials and methods. This has been undertaken by a
new international consortium, the Joint Committee for Traceability in Laboratory Medicine (JCTLM), formed in 2002. It brings
together experts representing the clinical laboratory profession, government agencies, and manufacturers, to promote international
comparability, reliability, and equivalence of measurement results in clinical laboratories for the purpose of improving healthcare.
Through the efforts of the JCTLM, manufacturers are able to assign values to kit calibrators with consistency using appropriate
higher order reference materials and methods, and traceability flowcharts, according to ISO Standards to ensure accuracy of test
results and to promote assay performance harmonisation. Users of assay kits can assess suitability of calibrators on the basis of
acceptable reference materials and/or methods identified by the JCTLM. The JCTLM exemplifies the dynamic nature of clinical
laboratory medicine, the inherent spirit of cooperation among professionals in this scientific field, and the international desire to
strive for the highest level of clinical laboratory practice for the benefit of patients.

Introduction providing the highest quality service feasible. Ideally, if two

Clinical laboratories exist to provide medical information or more laboratories at any location in the world tested the
for patient care. Day in and day out, the immediate product same patient specimen, equivalent values would be reported.
of laboratories are test results that are interpreted by health Global clinical laboratory practice has not yet reached this
care providers in light of the clinical presentation of each level of performance, but significant progress is currently
patient and in relation to the results of previous and future being made in this direction. The JCTLM is a significant part
values for the same measurand (analyte) and the values for of the movement towards assay standardisation and global
other measurands. Both medical care providers and patients harmonisation in the clinical laboratory community.
expect high quality service from every clinical laboratory.
The most important gauge of quality is that the test results Clinical Need for Standardisation
are accurate and suitable for medical practice. In fact, the It is undeniable that the clinical laboratory is experiencing
various customers of clinical laboratories expect (i.e. take globalisation. Physical geography has not changed; the world
for granted) that all test results produced by all laboratories is still round and considerable time and distance separates
at all times are accurate and clinically meaningful. So people. But yet it can be argued that the world is “shrinking”or
ingrained is this perception that medical errors, originating “flattening” when referring to the “virtual world” that has
in the laboratory or any other medical service, may be the been created by personal computers, the Internet, mobile
subject of front-page news, depending on how egregious the phones, and all of the other means by which individuals,
mistake. Clinical laboratories naturally pride themselves on organisations, and nations now communicate on a real time

Clin Biochem Rev Vol 28 August 2007 I 105

Armbruster D

basis with each other. Additionally, economies are more due to the fact that different assays exhibit non-equivalent
dependent on multiple countries to market their products. As a analytical responses, a spectrum of different values may result
result, people travel more, work in different countries and may from testing of the same specimen by different methods. This
find themselves seeking health care in different locations. In increases the likelihood that the result may be misinterpreted
addition, the need to “standardise” medical practice, including or that the results will be discordant with a previous value.
clinical laboratory practice, has increased in importance. The
implications of such globalisation are: As previously mentioned, PSA, one of the most common
tumour markers, provides another example of the lack of
1. An individual patient and/or his physician may find standardisation and the potential for very real impact on patient
himself with a test result obtained in one region or care. Currently, two sources of calibration are in common use
country and needing to compare that result with for PSA. One is based on the calibration scheme that produces
another result obtained in a different location. Without results consistent with the first PSA assay on the market. The
standardisation, the differences between the two results second is calibrated using the World Health Organization
may be uninterpretable. International Reference Preparation (IRP 96/670).5 Not only
2. Standardised clinical practice guidelines, in many are different results produced for the same patient specimen if
cases, dictate actions or treatments when a test result tested by assays using different calibration schemes, but clinical
is either greater or less than a given medical decision interpretation varies, with adverse consequences for patients. It
level. These levels are assumed to be independent of was found that 19% of patients were candidates for prostatic
the methodology used to obtain the result. For example, biopsy if analysed by a method based on the first calibration
further action may be taken when a cholesterol value is scheme but were not if tested by an assay based on the second
greater than 5.2 mmol/L or a prostate specific antigen calibration scheme, both sets of results having been assessed
(PSA) result exceeds 4.0 µg/L no matter what method against the traditional 4.0 µg/L cut-off. Without even knowing
is used to generate the value. In the case of cholesterol, which result is “right,” the reader can understand that either
the methods are sufficiently standardised to assure that biopsies may be performed when they are not warranted, based
reliable decisions are made. For PSA, however, that is on the PSA value, or biopsies will not be indicated when they
not the case. should be performed. Either scenario is obviously undesirable
and can result in an adverse patient outcome.
Other examples of measurands which may not be sufficiently
harmonised are creatinine, and HbA1c. These measurands The above examples (troponin, BNP and PSA) concern
are well-characterised compounds and current efforts to relatively complex, esoteric measurands for which
standardise results are underway. physicochemical analytical methods are not available and
that are tested using immunoassays. But calibration based on
Many other assays for key analytes, such as the cardiac reference materials and methods is critical even for very simple
markers troponin I and B-type natriuretic peptide (BNP), often elemental analytes, such as calcium.6 As recently reported, a
produce patently non-equivalent test results.1-4 Even a cursory manufacturer announced a restandardisation of its calcium
review of proficiency testing (PT) survey and external quality calibrators, based on atomic absorption.7 However, one
assurance (EQA) results demonstrate that reported values laboratory found that this resulted in a 5.4% change in value
can differ by orders of magnitude and are highly method assignment and exceeded the desirable total error for calcium
specific. PT samples, of course, may exhibit matrix effects of 2.4%. Further investigation comparing field method values
with some assays leading to discordant results, but even to the reference method, inductively coupled plasma-mass
comparison of fresh patient specimens confirms that radically spectrometry, and the Standard Reference Material (SRM) for
different values may be generated for the same measurand calcium developed by the National Institute of Standards and
by tests from different manufacturers. In the case of critical Technology, NIST SRM 956b, demonstrated that the change in
cardiac markers such as troponin and BNP, the lack of assay calcium kit calibrator values should have been much less. The
standardisation is not only inconvenient but a potential source repercussions of overestimation of calcium values are large.
of medical misinterpretation that could lead to patient safety It has been estimated that an analytical bias of only +/- 0.10
issues. Due to the lack of international method standardisation, mmol/L for calcium could increase the cost of patient care by
various assays require method-specific reference intervals, or approximately $150 million annually in the United States.6 In
“cut-offs”, as with some cardiac markers. This means that 2002 the Institute for Reference Materials and Measurements
physicians must interpret results relative to the appropriate (IRMM) conducted a study of the comparability of clinical
medical decision level for a specific assay. A patient specimen laboratory test results worldwide. Their objective was to
obviously contains a finite amount of any given analyte, but determine the level of comparability for 20 analytes including

106 I Clin Biochem Rev Vol 28 August 2007

The Joint Committee for Traceability in Laboratory Medicine

calcium.8 A graph in the IRMM report represents the average of traceability of test procedures have been available in the
calcium results obtained from 983 international laboratories. analytical community for several years. Several guidelines
From this graph, it is apparent many laboratories would have have been prepared describing both the advantages and the
been reporting patient calcium results with a bias greater than science underpinning traceability.9,10 Three important features
0.10 mmol/L. of traceability are:

There are a variety of reasons why assays for the same 1. A traceable method will have an unbroken chain from
measurand yield different values. The simple fact is that not a specific reference material and/or method to the final
all assays are created equal. The simple, small molecular result;
weight analytes, such as the electrolytes and glucose, tend to 2. An associated measurement uncertainty (an estimate of
exhibit a high degree of standardisation. This is because they the variation of the result) will be included; and
can be tested using well-developed, robust physicochemical 3. The methods will be validated and, where possible,
methods based on transferable, reproducible first principles, the commutability of each reference material in the
and pure reference materials are available. The more esoteric unbroken chain will be demonstrated.
analytes, such as those measured by immunoassays, are
more challenging and exhibit greater discordance. There are Traceable methods have several advantages. Many that apply
several factors preventing the standardisation of such analytes to the clinical laboratory are listed in Table 1.
including:
Table 1. Advantages of traceability of results in the clinical
1. Use of different calibrators by manufacturers because laboratory.
an internationally recognised reference material or
reference measurement procedure is not available;
2. Comparison of assays to different predicate devices • Universal agreement on nomenclature, definitions,
(competitor assays) or the use of inadequately qualified and assay performance goals
reference measurement procedures; • Ability of different assays for the same analyte to
3. Use of antibodies by manufacturers that recognise produce comparable results (i.e. quantitative results
different antigens or epitopes present on the same that are similar, although not necessarily identical)
analyte; and • Standard reference intervals as opposed to method-
4. Use of different capture or detection antibodies by specific intervals
manufacturers in two-step immunoassays for the same • Application of consistent standards of medical
analyte. care, best practice guidelines, Evidence Based
Medicine (EBM), and Laboratory Medicine Practice
Sometimes, differences among assays are due to chance Guidelines (LMPG)
but often the differences are intentional. Manufacturers are • Reduction of laboratory analytical errors and
restricted by patents from producing tests that are too similar enhanced patient safety
to other manufacturers, or manufacturers purposely seek to • Method comparisons to internationally accepted
set their assays apart from those of competitors. Whether reference measurement procedures instead of
intentional or not, many assays produce patient test results “predicate devices” (i.e. device performance is
that are not comparable and clearly not interchangeable. compared to “scientific truth” as determined by best
available method as opposed to performance of a
In summary, the Holy Grail for clinical laboratory medicine pre-existing device that may not represent the state
is the ability to generate the correct and “true” patient test of the art)
result on any specimen tested in any laboratory around the • Objective comparison of assays through Proficiency
world regardless of the specific analytical system used. The Testing/External Quality Assurance (PT/EQA)
patient test results provided by all clinical laboratories should surveys as opposed to relative performance
be accurate and meet medical interpretive needs, and the determined by peer groups
performance of the analytical systems that produce clinical
laboratory results should be comparable over space and time.
Method comparisons against reference measurement
Calibration Traceability and Standardisation procedures and validation of improved comparability through
One mechanism to improve the standardisation of laboratory proficiency testing/external quality assurance (PT/EQA)
test results is to assure that the methods are traceable to schemes deserve special mention. While it seems logical to
higher order reference materials and methods. The principles expect an IVD medical device to provide performance at

Clin Biochem Rev Vol 28 August 2007 I 107

Armbruster D

least equivalent to a device already available on-market, it is construct a suitable traceability chain. Instead, it points to
not always certain that the quality of the predicate device is ISO 17511 (In vitro diagnostic devices - Measurement of
adequate or representative of the “gold standard”. Equivalent quantities in biological samples - metrological traceability
performance between devices may not be adequate to ensure of values assigned to calibrators and control materials).13
that they are “fit for purpose” and meet clinical needs. A better ISO 17511 describes the elements of calibrator traceability
basis for comparison of routine “field tests” is a benchmark chains. Another useful document for those who are interested
analytical reference measurement system that has been in calibrator traceability is CLSI X5-R.14 This is a report
accepted as providing scientific “truth”. That is, a system that that provides guidance to IVD manufacturers in meeting the
represents the best analytical capability available at present. requirements of ISO 17511.

PT/EQA surveys were originally intended as educational ISO 17511

exercises to allow laboratories to compare themselves to ISO 17511 is based on metrology, the science of measurement,
their peers. While still useful for this purpose, PT surveys and provides a very useful roadmap for calibrator traceability.
now serve a regulatory purpose in many countries. There is The Figure illustrates a generic calibrator traceability
general agreement that target values for PT samples should be flowchart, with an unbroken chain linking the materials and
determined on the basis of reference methods and reference methods of the highest metrological order to the manufacturer’s
materials.11 This is the case for samples provided by some PT kit calibrators. The metrological scheme described in ISO
survey providers, but “peer group” grading is still common. 17511 assumes that appropriate reference materials and
The peer group approach allows a laboratory to gauge whether reference measurement procedures exist for measurands. This
its performance is comparable to other labs using the same is not always the case. In fact, five different scenarios exist:
method/analyser. However, peer group comparison leaves
open the question of the absolute accuracy of a test method 1. Reference material and reference measurement
as there can be multiple “true values” (i.e. each peer group procedure are both available and traceable to SI units
mean value represents a relative “true value”). Comparison (the ideal; e.g. glucose, creatinine);
to the “true value” as determined by a reference measurement 2. Reference material and reference measurement
procedure allows both an absolute and relative performance procedure are both available but not traceable to SI;
yardstick. Although uniform calibrator traceability by 3. Reference material is not available, but a reference
IVD manufacturers is no guarantee of equivalent assay measurement procedure is available (e.g. coagulation
performance, it definitely promotes assay standardisation and factors);
minimises the difference between method peer groups. 4. Reference material is available, but a reference
measurement procedure is not available (e.g. specific
The In Vitro Diagnostics Directive and Traceability plasma proteins such as transferrin, immunoglobulins);
The In Vitro Diagnostics Directive (IVDD) in Europe and
represents a major step forward in promoting assay 5. Neither a reference material nor reference measurement
standardisation and global harmonisation in the clinical procedure is available (e.g. some tumour markers).
laboratory community.12 The IVDD became effective on
7 December 2003, and applies to all manufacturers who In the last situation, manufacturers must prepare calibrators
provide calibrators for IVD devices in the European Union. from the best options available, increasing the chances that kit
Compliance with the IVDD allows manufacturers to affix calibrators will be traceable to very different starting materials
the “CE” mark on their products as an assurance of quality. and unique to a given manufacturer’s assay.
Without the CE mark, IVD devices cannot be sold in the
European Community. The IVDD requires manufacturers to Caveats about the IVDD and Metrological Traceability
provide information about the traceability of their calibrators The IVDD does not identify internationally recognised
when requested, including estimates of uncertainty for the reference materials and measurement procedures.
values assigned as appropriate. This entails constructing Manufacturers are expected to select the appropriate reference
“traceability chains” that begin with reference materials and/ systems to anchor their calibrators, judiciously choosing from
or reference methods of the highest metrological order and to those that are available. Nor does the IVDD require that new
document an unbroken link from the reference materials and/ or improved reference measurement systems be developed
or methods to the kit calibrators used in clinical laboratories. for calibrator traceability and be applied by manufacturers.
The IVDD requires that calibrator traceability information be Naturally, it is expected that improved reference materials
made available for Competent Authorities, Notified Bodies, and methods will be developed over time resulting in
and users of IVD devices, but it does not describe how to changes to calibrator traceability and improved accuracy and

108 I Clin Biochem Rev Vol 28 August 2007

The Joint Committee for Traceability in Laboratory Medicine

Figure. A generic calibrator traceability flowchart. Based on ISO 17511:2003 Fig 4.2.2.

standardisation of assays. The reality is that, if the materials in close agreement when used to test a candidate reference
and methods used to develop the unbroken chain are not material. For example, a specification for commutability
well designed, the traceability scheme may not provide the might be that test results agree within a range representing
expected improvement in comparability of results. Metrology 95% of the patient’s results in a comparison study. Fresh
allows for multiple traceability schemes and leaves it up patient specimens (healthy and diseased individuals) should be
to the user to develop and validate the calibration scheme tested by field methods in addition to the reference material to
that is technically best for the manufacturer’s product(s). demonstrate that the reference material mimics the analytical
The problem then becomes, how accurate are the materials, response of patient specimens.
methods and laboratories being used to assign values to the
manufacturer’s working calibrators. This is where using The Joint Committee for Traceability in Laboratory
materials and methods of a “higher metrological” order to Medicine
construct an unbroken link to kit calibrators, should help in It is clear that the key to assay standardisation and the global
generating results that are close to “the true value.” However, harmonisation of clinical laboratory test results rests with the
unless the materials and methods are of an appropriately practical application of metrological concepts and the use of
higher order, two or more “metrologically legitimate” internationally accepted reference materials and methods. But
traceability chains can be constructed for some measurands who is the arbiter of traceability in the clinical laboratory field?
and each can produce “a true value” that may or may not As noted by Stenman: “Although quite a few organisations
be equivalent or even quantitatively close, but nevertheless deal with standardisation, it is not clear who is responsible
are considered to be metrologically acceptable. Of course, for what. Because standardisation is an international rather
what is desired is a traceability chain that produces “the true than a national or regional problem, it is desirable that
value”- a value that is generated by a sufficiently accurate one international organisation should be responsible for
measurement and that represents “scientific truth” (or as close the coordination of various standardisation projects.”15 In
to absolute scientific truth as current technology allows) as Stenman’s view, “Currently, most of the standardisation
opposed to multiple forms of “relative truth” (results that are problems are taken care of by assay manufacturers, and
valid according to the traceability chains on which they are this situation will not change soon.” Indeed, the premise of
based, but not quantitatively equivalent). While the IVDD the IVDD is that manufacturers will play the primary role
is a great leap forward for global harmonisation of clinical in traceability. Manufacturers are the logical leaders of the
laboratory practice, manufacturers and users of IVD devices traceability/global standardisation movement, but how will
must consciously strive to ensure metrological consistency by they know which reference materials and methods they
the use of internationally recognised and accepted reference should use to anchor their assays? Of necessity, the JCTLM
systems. was created.

Another complication is that reference materials need to be The JCTLM was formed in 2002 and is an international
“commutable.” That is, they should produce an analytical consortium sponsored by the Bureau International des
response with a reference measurement procedure and Poids et Mesures (BIPM, or the International Bureaus of
with routine field methods mimicking that of fresh patient Weights and Measures), the International Federation of
specimens. All methods should generate results that are Clinical Chemistry and Laboratory Medicine (IFCC), and the

Clin Biochem Rev Vol 28 August 2007 I 109

Armbruster D

International Laboratory Accreditation Cooperation (ILAC), Each of these specialised subgroups is headed by a Review
bringing together government agencies, the clinical laboratory Team Leader and typically consists of four to eight Team
profession, and industry. The goal of the JCTLM is to support Members. Team Members are drawn from all around the
worldwide comparability, reliability, and equivalence of world, chosen for their expertise, and represent NMIs,
measurement results in clinical laboratories for the purpose governmental regulatory agencies, IVD manufacturers, and
of improving healthcare.16 The information provided by the individuals working in the clinical laboratory professions
JCTLM is intended for the IVD industry and any individual (e.g. academia, hospitals, accreditation bodies). Significant
or organisation dealing with traceability of clinical laboratory attempts are made to assure that the balance of government
medicine purposes. This includes the suppliers of PT/EQA agencies, IVD manufacturers and clinical laboratory personnel
programs. is maintained, along with a geographic balance. Any person
with the required expertise and the desire to participate in the
Among the stakeholders invited to apply for JCTLM JCTLM’s activities is encouraged to volunteer; application
membership are: intergovernmental organisations; national and disclosure of interests forms are available from JCTLM
governmental organisations such as national metrology website.17
institutes (NMIs); and any international, regional, and national
non-governmental organisations with technical competency At the onset, the JCTLM recognised that it required a
in the field. Member organisations may send representatives transparent process based on internationally agreed upon
to JCTLM meetings, have access to the documents created by standards to identify what methods, materials and laboratories
the JCTLM, participate in their working groups by nominating adhere to the definition of “higher order”. To meet this need,
members to them, and submit written statements concerning the JCTLM abides by two main principles:
matters under consideration by the JCTLM.
1. Both working groups are guided by the following ISO
The JCTLM consists of two working groups. Working Group standards
1 is responsible for identifying internationally accepted • Mainly ISO 17511: In vitro diagnostic medical devices
reference measurement procedures and reference materials. – Measurement of quantities in biological samples
Working Group 2 is responsible for identifying reference – Metrological traceability of values assigned to
laboratories that provide internationally accepted reference calibrators and control materials;
measurement procedures to be used for the value assignment of • ISO 15193: Measurement of quantities in samples
calibrators and the validation of commutability of calibrators of biological origin – presentation of reference
and other reference materials. measurement procedures;
• ISO 15194: Measurement of quantities in samples of
Both working groups utilise a series of review teams to biological origin – description of reference materials;
provide the resources and expertise to identify acceptable • ISO 15195: Laboratory medicine – Requirements for
reference materials, methods and laboratories. These review reference measurement laboratories; and
teams encompass a range of analytes as listed below. • ISO 18153: In vitro diagnostic medical devices
– Measurement quantities in samples of biological
• Blood Gases origin – Metrological traceability of values for catalytic
• Blood Groupings concentration of enzymes assigned to calibrators and
• Coagulation Factors control materials.
• Drugs
• Electrolytes These provide the internationally agreed upon requirements
• Enzymes for what constitutes higher-order reference materials and
• Metabolites-Substrates methods.
• Microbiology Serology
• Non-Electrolyte Metals 2. JCTLM Quality Manuals describe in detail the activities of
• Non-Peptide Hormones the working groups and review teams. These quality manuals
• Nucleic Acids can be found and downloaded from the JCTLM web pages.18,19
• Proteins These quality manuals provide the transparency of the
• Quality Systems decision-making processes used by making the public aware
• Vitamins of the practices and procedures used to make the decisions.

110 I Clin Biochem Rev Vol 28 August 2007

The Joint Committee for Traceability in Laboratory Medicine

There is also an Implementation Protocols Team under Limitations of the JCTLM Process
Working Group 1. This team’s mission is to address the use Naturally, the JCTLM faces several challenges. Like many
of reference materials, reference measurement procedures, similar professional organisations, it depends on volunteers
and reference laboratories by the IVD industry. Specific tasks and their expertise. Many laboratory professionals are actively
include: engaged in JCTLM activities, and certainly the organisations
to which these individuals belong (NMIs, governmental
1. Suggesting a realistic timeline for manufacturers to regulatory agencies, medical and graduate schools and
re-standardise assays using traceability to appropriate other academic facilities, medical centres, hospitals, and
reference materials and methods; manufacturers) support this involvement. But the reality
2. Recommending efficient means to make the JCTLM is that JCTLM participation is for most of the volunteers
information available to the industry (along with an “extracurricular activity”, effort that is meaningful and
changes as they occur); necessary, but which typically is not part of the “day job”
3. Providing a mechanism for reviewing and approving and may not be a top priority for employers. A real concern
new candidate reference materials and methods as they is whether companies, government bodies, medical centres,
become available; and and other employers will support the volunteers’ efforts and
4. Making available an effective forum for communication allow them to contribute time and effort. Another potential
of traceability and standardisation information among weakness is that the NMIs and other institutions that typically
the members of the IVD manufacturing community. provide the necessary reference materials and methods may
have limited resources to maintain the necessary metrological
The self-appointed tasks of the JCTLM are daunting. It is infrastructure to meet the needs of the global clinical laboratory
estimated that clinical laboratories perform analyses for community. It is not unusual for NMIs to be subject to budget
about 400–1000 different measurands. Only about 10% of constraints from their respective governments. Even while
these measurands are what the JCTLM describes as type A recognising the logic of traceability, standardisation, and
(JCTLM List 1) analytes, those for which well-recognised global harmonisation efforts, many stakeholders may have
reference materials and methods exist and can be traced to difficulties allocating the human and other resources to
the SI unit. Some of the measurands have either a reference support the JCTLM.
measurement procedure or a reference material but not both.
Most of the measurands which have references that meet the In addition, the JCTLM operates by consensus. Obtaining
criteria of “higher order” are the common, routine clinical consensus among the members may, in some cases, be easy,
chemistry tests, such as the electrolytes (Na, K, Cl), glucose but in others it may require considerable discussion. After all,
and cholesterol. These are examples of type A analytes. But if agreement on internationally accepted reference materials
about 80% of analytes fall into the type B (JCTLM List 2) and methods were simple, there would have been no need to
analytes category. Type B analytes consist of the more esoteric form the JCTLM in the first place. Furthermore, there is not
tests, such as coagulation factors, tests for nuclear materials, a fixed JCTLM budget. The participating organisations are
and immunoassays, (including those for hormones, cardiac expected to fund activities on a “pay as you go” basis.
markers, tumour markers, vitamins, and viral markers). In
the clinical laboratory about 80% of total lab tests represent Issues for Manufacturers
type A analytes and about 20% represent type B analytes. To As noted previously, the primary onus is on the manufacturers
date, the JCTLM has identified reference materials and/or to drive traceability, and they face some major challenges.
methods for about 120–150 analytes. The fact of the matter While the JCTLM will identify appropriate references when
is that many analytes lack higher order reference materials, they are nominated and available, for many analytes they
reference measurement procedures, or both. The goal of the simply do not exist. The manufacturers must then decide how
JCTLM is not to provide reference materials and methods, to best “anchor” their assays, i.e. make them traceable, given
only to qualify candidate materials and methods as acceptable the available options. Reference materials may be available
for traceability purposes and make this information available for some measurands, but not others. For example, a serum/
to the global clinical laboratory community. Knowing that plasma reference material may exist for an analyte, but a urine
a reference material method or laboratory service is not or whole blood based reference material may not exist for the
available, or does not meet the ISO criterion of a higher order same analyte. In some cases, as for the enzymes, reference
reference is also useful information for manufacturers and preparations may be available, but they may not be human
clinical laboratories. sourced materials and may consist of a single isoenzyme
although two or more isoenzymes are present in a patient
specimen. Even when very pure reference materials can be

Clin Biochem Rev Vol 28 August 2007 I 111

Armbruster D

obtained, they may not be commutable, thus limiting their true for the more esoteric, type B measurands that are analysed
effectiveness and the degree of agreement between methods using immunoassays. The very nature of many of these
they can provide. Reference measurement procedures also measurands makes it difficult for different methods to produce
pose challenges as they are quite demanding and require a comparable results. Nevertheless, the closer comparability of
high level of discipline. For example, isotope dilution gas immunoassays is a worthy and practical goal.
chromatography/mass spectrometry reference methods
are typically available from only a handful of reference It should be recognised that the application of metrology
laboratories and are not readily set up in manufacturers’ and traceability alone does not necessarily equate with good
facilities. There are many issues faced by manufacturers when clinical laboratory science. The often accepted assumptions
developing new assays and the main ones are listed below. expected of metrology are that the method being used is
“perfect” in that there will be no matrix effects or individual
• Reference materials not expressed in SI units sample variations and the methods and materials are highly
(e.g. human chorionic gonadotrophin) precise giving no bias. As a result, metrology alone will work
• Time required for restandardisation of an assay when the methods are highly specific and part of a more “pure”
(e.g. 18-24 months) and refined measurement system but may not be as effective
• Restandardisation when a new reference material or on complex matrices such as the patient samples used in the
measurement procedure is recognised clinical laboratory. Designing assays that generate equivalent
• Satisfying regulatory requirements of different countries results and values that are as accurate as possible and as close
and global regions as feasible to absolute scientific “truth” is, however, clearly
• Producing assays that will meet the different standards desirable. However, this does not negate the excellent work
of care and medical practice around the world that clinical laboratories currently do. The fact is that clinical
• Development of internal references when no others exist laboratories have been making vital contributions to medicine
for decades even though assays for some measurands may
Currently manufacturers use the information supplied by the produce widely divergent results and values may be reported
JCTLM to accomplish the following: in different units (e.g. SI vs conventional units). Clinical
laboratories can fulfil the majority of their patient care mission
1. Provide calibrator traceability information to clinical even if every lab throughout the world doesn’t produce
laboratories and regulatory bodies; equivalent results for the same patient specimen regardless
2. Restandardise assays, making them traceable to of the measurement system used. Effective and clinically
internationally accepted reference materials and meaningful laboratory operations in support of the practice
methods; of medicine trump the rigorous adaptation of metrology to
3. Improve manufacturing procedures to decrease lot to lot the clinical laboratory. However, global standardisation
variability of calibrators and reagents; efforts can contribute to improved patient care and foster
4. Support professional organisations involved in global closer laboratory-clinician interactions as exemplified
standardisation activities (e.g. JCTLM, American by the estimated glomerular filtration rate (eGFR) and
Association for Clinical Chemistry, IFCC, Australasian standardisation of serum creatinine measurement.
Association of Clinical Biochemists, ISO, CLSI); and
5. Design assays to meet medically relevant total error, The JCTLM, regulatory bodies, professional societies,
imprecision, and bias goals. and IVD manufacturers all have key roles in promoting
calibrator traceability, assay standardisation, and global
The Future of Global Standardisation in the Clinical harmonisation. But individual clinical laboratories also have
Laboratory the responsibility to select field methods that demonstrate
It can be expected that the application of metrological concepts metrological traceability and that meet the clinical standards
in the clinical laboratory will continue in the 21st century, suggested by laboratory medicine practice guidelines. Labs
certainly affecting clinical chemistry but impacting the other must also continuously assess their routine methods to ensure
specialties within the field as well. The obvious, and good, that they consistently produce accurate, medically useful (“fit
reasons for this trend have previously been explained. There for purpose”) results. Thus there is an onus on laboratory
are some caveats that should also be noted. The identification directors and clinical laboratory scientists to understand
of reference materials and methods by the JCTLM and their the significance of reference materials and methods and
use to properly document traceability of calibrators will how they should be used to guarantee the quality of assays.
promote assay standardisation and global harmonisation of
patient test results but will not guarantee it. This is especially

112 I Clin Biochem Rev Vol 28 August 2007

The Joint Committee for Traceability in Laboratory Medicine

Conclusions 8. Van Nevel L, Örnemark U, Smeyers P, Harper C,

Various imperatives in the global clinical laboratory Taylor PDP. IMEP-17 Trace and Minor Constituents in
community, including the medical need for standardisation Human Serum EUR 20657 EN Report to Participants.
of assays and the regulatory requirements of the IVDD to Part 1: International comparability. Retiesweg,
document metrological traceability, are currently impacting Belgium: Institute for Reference Materials
the practice of clinical laboratory medicine. It is to be and Measurements. http://www.irmm.jrc.be/html/
expected that the movement to firmly establish proper and interlaboratory_comparisons/imep/imep-17/IMEP17_
internationally accepted traceability chains for assay kit report_part1.pdf
calibrators, allowing them to be linked by an unbroken chain 9. EURACHEM / CITAC. Traceability in Chemical
to reference materials and methods of the highest metrological
Measurement. A guide to achieving comparable results
order, will continue indefinitely. This trend in turn will
in chemical measurement. 2003. http://www.citac.cc/
promote assay standardisation and global harmonisation
EC_Trace_2003.pdf
in the clinical laboratories of the 21st century. The pace,
completeness, and success of this movement depends on 10. EURACHEM/CITAC. Quantifying uncertainty in
cooperation among a variety of entities, including professional analytical measurement. Second Edition. 2000. http://
societies, regulatory bodies and other governmental www.citac.cc/QUAM2000-1.pdf
agencies, industry, and individual clinical laboratories. 11. Rej R, Norton-Wenzel CS, Cao Z. Target values and
The JCTLM is a new organisation, just formed in 2002, method evaluation in proficiency testing programs. Clin
but it can be expected to play a central role in these efforts. Chem 2001;47:2185-6.
12. Greenberg N. Calibrator traceability: the industry
Competing Interests: None declared.
impact of the IVD Directive’s new requirements. IVD
Technology 2001;7:18-27.
References
1. Panteghini M, Gerhardt W, Apple FS, Dati F, Ravkilde 13. ISO 17511:2003. In vitro diagnostic medical devices
J, Wu AH. Quality specifications for cardiac troponin -- Measurement of quantities in biological samples --
assays. Clin Chem Lab Med 2001;39:175-9. Metrological traceability of values assigned to calibrators
2. Apple FS. Standardization of cardiac markers. Scan J and control materials. ISO, Geneva, Switzerland.
Clin Lab Invest Suppl 2005; 240:107-11. 14. Clinical and Laboratory Standards Institute. Metrological
3. Apple FS, Panteghini M, Ravkilde J, Mair J, Wu AH, traceability and its implementation. CLSI document X5-
Tate J, et al. Quality specifications for B-type natriuretic R. Wayne, PA: CLSI. 2006.
peptide assays. Clin Chem 2005;51:486-93. 15. Stenman UH. Immunoassay standardization: is it
4. Christenson RH, Duh SH, Apple FS, Bodor GS, Bunk possible, who is responsible, who is capable? Clin Chem
DM, Panteghini M, et al. Toward standardization of 2001;47:815-20.
cardiac troponin I measurements part II: assessing 16. Mission statement for Joint Committee on Traceability
commutability of candidate reference materials and in Laboratory Medicine. http://www.bipm.org/cc/
harmonization of cardiac troponin I assays. Clin Chem JCTLM/Allowed/Workshop/JCTLMMIssion13June.
2006;52:1685-92. pdf (Accessed 29 June 2007).
5. Link RE, Shariat SF, Nguyen CV, Farr A, Weinberg AD, 17. JCTLM: Joint Committee for Traceability in Laboratory
Morton RA, et al. Variation in prostate specific antigen Medicine. http://www.bipm.org/en/committees/jc/jctlm/
results from 2 different assay platforms: clinical impact (Accessed 29 June 2007).
on 2304 patients undergoing prostate cancer screening.
18. JCTLM-WG1 Quality Manual: Reference Materials
J Urol 2004;171:2234-8.
and Reference Procedures. http://www.bipm.org/en/
6. National Institute of Standards and Technology. Planning committees/jc/jctlm/jctlm-wg1/wg1_quality-manual.
Report 04.1. The impact of calibration error in medical html (Accessed 29 June 2007).
decision making. 2004. http://www.nist.gov/director/
19. JCTLM-WG2 Procedure Manual: Reference
prog-ofc/report04-1.pdf
Measurement Services. http://www.bipm.org/en/
7. Bais R. What information should manufacturers provide
committees/jc/jctlm/jctlm-wg2/wg2_quality-manual.
on their procedures? Clin Chem 2006;52:1624-5.
html (Accessed 29 June 2007).

Clin Biochem Rev Vol 28 August 2007 I 113