An overview of EFLM harmonization activities in Europe

Working Group: Biological Variation (WG-BV)

Biological variation (BV) is a fundamental source of variation, and although this does not lend itself to harmonization per se, there is a need to establish as accurate an assessment of within- and between-subject BV as possible for each of the test analytes we currently measure. This is not just for academic reasons but is fundamental to developing harmonized analytical performance specifications (APS) for individual tests. The 1st Strategic Conference of the EFLM on “Defining analytical performance goals 15 years after the Stockholm Conference on Quality Specifications in Laboratory Medicine” was held in Milan in 2014, and from it a consensus was agreed whereby the analytical goals for a particular measurand should be based on a hierarchy of models [2]. Ideally, the APS for a test should be based on the model of clinical outcome studies. However, such data often do not exist for many laboratory tests and to use this approach there should also be a strong link between the measurand and the clinical situation. Because many measurands in the laboratory are used for multiple purposes, this is difficult to achieve. However, it should be possible for some measurands such as HbA1c and CRP. In the laboratory, there is often a reliance on the next preferred model that is based on BV, whose use attempts to minimize the ratio of “analytical noise” to the biological signal. The rationale for using APS based on BV is that such measurands should have a stable homeostatic set point and preferable be regulated in a strict way. Examples are electrolytes such as sodium and potassium [3].

The same Strategic Conference identified the main limitations of BV data that existed at that time. One of these was the fact that there was no standardized method for assessing BV and, as a consequence, studies published could be of variable quality. However, this also meant that there was still a potential to assess the BV of many test analytes to as high a standard as was possible using contemporary statistics and study designs.

These issues gave a continued focus to the EFLM’s WG: BV, and its work has recently been coming to fruition. Building on an original checklist [4], they have created the BV Data Critical Appraisal Checklist (BIVAC) that can be used to evaluate to the quality of existing BV studies [5]. They have also embarked on an extensive Europe-wide project to determine the BV of numerous tests using rigorous protocols [6]. Called the European Biological Variation Study (EuBIVAS), this has already published BV estimates for many routinely measured enzymes [7] and for creatinine [8], with plans for reporting on many other tests, including those of lipids, electrolytes and coagulation [9]. Perhaps not surprisingly, their findings to date suggest that both within- and between-subject BV tends to be tighter when their exacting procedures are followed [9].

These data will also be included in a database of BV studies, which will update the seminal work of Ricós et al. [10], and it is planned to soon be hosted by the EFLM on its website.

Working Group: Preanalytical Phase (WG-PRE)

The preanalytical phase is known to currently be a large source of error in the patient testing process [11]. Part of reason for this is likely to be the variability in practice that exists throughout Europe. The preanalytical WG of the EFLM has been active in recommending the harmonization of many areas of this activity which are of immediate relevance to the work of most laboratories [12].

Taking the WG’s achievements in the order in which a patient would have a sample taken, they have recommended how a patient who requires a fasting sample should be prepared prior to sample collection [13]. Validating or verifying laboratory equipment and reagent lots is now commonplace, but the same is not true for blood containers, so the WG-PRE has provided advice on how this can also be achieved locally [14]. They have also published recommendations to help minimize the perennial issue of patient and blood container identification [15] and published a systematic review on tube additives in blood glucose measurement [16]. The process of venous blood collection itself is poorly harmonized, so the WG have conducted two European surveys to understand how this varies currently and are in the process of using the findings to write recommendations on how this should be performed more uniformly. The color coding of the blood containers themselves varies between laboratories, and tube manufacturers and are an obvious potential source for confusion amongst users. The WG-PRE’s associated TFG on Color Coding has seen this as a priority and has recommended the harmonization of container tube colors and their associated labels [17]. The need for a specific sequence in which different blood tubes should be collected (the “order of draw”) has been a long-standing topic of discussion, especially in an era of evacuated collection containers. The WG-PRE now provides their view that an order of draw should still be followed, primarily because it is not always possible to ensure that every blood collection will follow the ideal process [18]. Recently, they have also provided practical recommendations for the management of hemolyzed samples [19].

Beyond publications, their harmonization initiatives have also formed a basis for four European conferences on the preanalytical phase, which has given rise to several opinion papers [20], [21], [22], [23].

Working Group: Guidelines (WG-G)

Guidelines, of course, go hand in hand with practice harmonization. The National Societies, which comprise the EFLM, already have many laboratory-related guidelines that are of excellent quality, and one of the tasks of the WG-G is to collect these into the repository that is currently found on the EFLM website [24]. Some of these guidelines are so comprehensive that for the EFLM to reinvent them would, at best, be time wasteful, so it is the intention of the group that some of these could be adapted or adopted as European guidelines (with the National Society’s permission) to avoid any duplication.

Clinical practice guidelines are critically examined by the WG-G according to a checklist of essential information required for use of biomarkers [25]. In case the WG-G identifies key information that is missing (or inconsistent) for the correct use of a biomarker, e.g. preanalytical, analytical or postanalytical requirements, an EFLM guideline will be produced by an expert group to address and harmonize the critical issues.

The plethora of guidelines on similar topics can also be rationalized by having joint guidelines with other clinical societies. A successful example of this cooperation is the joint consensus statement between the EFLM and the European Atherosclerosis Society on lipid determination on non-fasting samples [26].

Working Group: Harmonization of the Total Testing Process (WG-H)

As its name suggests, this WG has harmonization of practice at the heart of its purpose. It has surveyed the National Societies of the EFLM on aspects of the testing process such as consistency in test profiles, specific disease guidelines, autovalidation rules and the thorny issue of what test units are used in their country. With respect to the last of these, small steps are being taken, such as the Group recommending that the denominator unit for test reporting be per liter (/L) rather than, say, per milliliter (/mL) by default [27]. This leads to a change closer to the SI unit (such as from mg/mL to g/L) but will not affect the numeric result. It is planned that the EFLM will, in due course, issue a recommendation of its preferred reporting units for tests to form the basis for adoption. The group is also exploring means of harmonizing reference intervals, where possible.

Working Group: Postanalytical phase (WG-POST)

This WG has concentrated on the value, or otherwise, of adding interpretative comments onto laboratory reports. They have surveyed how clinicians or laboratory specialists interpret the results of key laboratory tests in clinical decision making in different clinical scenarios throughout Europe [28], [29], [30] and have performed studies about laboratories’ practices and policies in postanalytical phase, including interpretative commenting [28]. They have established a collaboration with colleagues in the International Federation of Clinical Biochemistry and Laboratory Medicine (IFCC) Working Group on Harmonization of Interpretative Comments EQA to have a joint approach to harmonization of interpretative commenting globally and developing joint recommendations on this subject.

Working Group: Patient Focused Laboratory Medicine (WG-PFLM)

The main objective of this WG is promoting patient engagement with Specialists in Laboratory Medicine. This is currently being addressed through surveying the views of patients receiving their own laboratory results and who might be best placed to communicate them [31] and how, which has implications for harmonization; the Group is also embarking on a project that should ultimately help harmonize the criteria to judge the acceptability of mobile phone and computer apps that incorporate laboratory medicine data.

Working Group: Test Evaluation (WG-TE)

As potential new diagnostic tests become available, there is a need to evaluate their utility in a uniform way [32]. The WG-TE has therefore written a practical guide to evaluating new biomarkers for this purpose in a harmonized way [33].

Laboratory medicine has a poor record bringing new tests to market in a timely and effective way. Evidence-based laboratory medicine (EBLM) provides the underlying principles for how a new biomarker should go through the test evaluation process. An interactive Test Evaluation course has been set up by the WG-TE that aims to address that gap by extending the principles of EBLM to provide some practical tools for the key processes of test evaluation. Tools to conduct test evaluation include identifying unmet needs for biomarkers, clinical pathway mapping to define the role and purpose of biomarkers in changing outcomes, determination of analytical and clinical performance characteristics of biomarkers and assessing the clinical effectiveness of biomarkers. The course given in Leiden in 2016 was repeated in Sydney in 2018.

Working Group: Personalized Laboratory Medicine (WG-PLM)

One of the main objectives of this group is to establish how the fledgling of proteomics, metabolomics and pharmacogenetics can be incorporated into the laboratory medicine discipline [34]. From a harmonization perspective, they have helped develop the first preanalytical assessment of quality in molecular diagnostics [35].

New Task Groups for APS

APS in laboratory medicine should ideally be harmonized at least within certain geographical areas and for specific situations. The 1st EFLM Strategic Conference in 2014 dealt with APS, both because it was necessary to set the theoretical framework for how to set performance specifications and also to try to harmonize performance specifications in special situations. A consensus document stating that performance specifications could be based on three different models – namely clinical outcome, BV and state of the art – was developed [2]. After the conference, a task force comprising five different TFGs were established to deal with different aspects of performance specifications. The main results from these groups have already been published [36]. The TFGs are now closed, but some of the ideas will be taken forward in two new TGs. The TG-Biological Variation Database, located within the WG-biological variation, will establish a database including APS for different measurands based on BV. The other TG on Performance Specifications based on Outcome Studies (TG-PSOS), within the WG-Test Evaluation, will establish the theoretical and practical basis for how to set performance specifications based on outcome studies. The intention is that a list of APS for different measurands shall be produced. The two TGs shall cooperate to allocate different measurands to the different models for setting APS and to develop APS that can be proposed for EQA organizations.