Abstract for the EQALM post-congress day, Prague, June 1^st 2001:

Classification codes for laboratory tests (NPU codes) ……..……………… 37

Design of Reports in External Quality Assessment ………………………… 39

Software for Data Managing and Publishing og EQAS in Cuba ………….. 41

2^nd International Standard for Fibrinogen in Plasma in the Standardisation

of the Analysis of the Concentration of Fibrinogen in Plasma, and the Ci-

trate Dilution of Patients Plasma …..……………….………………………… 43

Abstract for the EQALM post-congress day, Prague, June 1^st 2001

Organizers of external quality assurance (EQA) surveys often operate on a regional basis, which makes it easy to adapt to local traditions used for terminology and classification in the field of laboratory medicine. The type of information collected from each participant in an EQA survey also varies in accordance with regional habits including type of calibration, reagent/ procedure and instrument.

The value of the results of EQA surveys should increase if comparisons and meta analysis of corresponding data from different EQA organisers were possible. Such comparisons necessitates not only a common way to use statistics – such as principles for outlier detection – but also a common nomenclature for the tests performed, calibrators and methods used by the participants. If EQA data were to be handled in a more standardized manner than today, we could also benefit from easier development of compatible software solutions for data-analysis and for electronic data transmission. The use of a strict nomenclature for measured properties would in addition facilitate the work for standardization of laboratory methods and the use of common reference intervals. Currently, the nomenclature of parameters intended to be measured in a EQA survey varies. Some EQA providers use only the name of the component, e.g Glucose, to identify the property which is expected to be measured in a sample. If not clear from the context, the full name of the property, e.g Plasma—Glucose; substance concentration, ought to be used, to distinguish from other properties such as Blood--Glucose or Urine--Glucose. As the measured properties that are dealt with by medical EQA organizers are identical with those of interest for the medical practice, each property can be defined using the nomenclature system developed for clinical laboratory investigations by the IFCC/IUPAC joint committee on Nomenclature and Properties (C-NPU). The NPU system covers measured property fully traceable to Systeme International (SI), or to a calibrator recognized by WHO or comparable metrological level, together with a unique NPU-code. The code for Plasma—Glucose is thus NPU02192 and the code for Blood—Glucose is NPU02187. The NPU system is now fully implemented in medical laboratories in Denmark and partly in Sweden.

In addition to the description of the measured property the EQA organizer has to collect information on general procedures including calibration, equipment used and reagent lot. Until now no generally accepted nomenclature system has existed for these types of data. Therefore all EQA-organisers have developed various in-house systems to categorize the results produced, e.g in the Finnish Labquality organisation, where a useful set of codes for methods, instruments and reagents traceable to lot number have been developed. A global medical device nomenclature (GMDN) has recently been developed, based on the structure in ISO 15225:2000: Nomenclature – Specification for a nomenclature system for medical devices for the purpose of regulatory data exchange. The intended users are, according to the ISO document, not primarily organizers of external quality assurance, but notified bodies and health authorities. So far the content of the database has not been evaluated, and a question that still remains to be answered is whether this GMDN will fulfil the requirements for a coding scheme to be used by the EQA organizers. As EQA organizers, we have to organize very specific information on reagents, and calibrators, types of instruments and measurement procedures. The GMDN might be the final solution; if not, we have to develop a nomenclature system for our purpose.

The participation of clinical laboratories in External Quality Assessment Schemes (EQAS) is a key element in the evaluation and improvement of testing procedures and is considered mandatory for accreditation. Therefore, it is important for laboratories to participate in EQAS complying to formally recognised requirements. According to ISO/FDIS 15189 (1) and ISO/IEC 17025 (2) the laboratory shall participate in inter-laboratory comparisons that substantially are in agreement with ISO Guide 43-1 (3).

Clinical laboratories have a long tradition of participation in External Quality Assessment Schemes (EQAS), and numerous results that have been achieved in EQAS are reported and published. However, almost all EQAS use different statistical evaluation procedures and different designs of reports.

The content of EQAS reports will vary depending on the purpose of the particular scheme, but according to ISO Guide 43-1 (3) and ILAC–G13: 2000 (4) it is possible to specify the essential principals that need to be considered when designing a report. Each report should be clear and comprehensive and include data on the distribution of results from all laboratories together with an indication of individual participant performance.

The general requirements when designing a report in accordance with ISO Guide 43-1 (3) and ILAC–G13: 2000 (4) are as follows:

Each scheme should include names and affiliations of persons involved in the design and conduct of the scheme, number and date of issue of report, clear description of items or materials used, including details of sample preparation and homogeneity testing, laboratory participation codes and test results, statistical data and summaries, including assigned values and range of acceptable results, procedures used to establish any assigned value, details of the traceability and uncertainty of any assigned value, comments on laboratory performance by the co-ordinator and technical advisers, procedures used to design and implement the scheme (e.g. reference too a scheme protocol), procedures used to statistically analyse the data and interpretation of the statistical analysis. For schemes operated on a regular basis, it may be sufficient to have simpler reports such that many of the requirements could be excluded from routine reports, but included in periodic, e.g. annual, summary reports and on request from participants.

That the requirements are general give the opportunity for renewal and development, but, on the other hand, the requirements can be interpreted differently, resulting in dissimilar reports, reflected both in modifications of terminology and in recommendations on how to interpret and present results from participants. Where there is more than one EQAS for the same parameter with different designs of reports, this could cause problems.

The modification of the terminology could lead to misunderstandings and different actions to be made among the participants, mainly concerning the comments on laboratory performance by the co-ordinator and technical advisers. The various statistical techniques used could also lead to misinterpretation. ISO Guide 43-1 (3) has useful examples of statistical methods for the treatment of EQAS data, but should be supplemented by ISO/DIS 13528 (5) for guidance on the selection of statistical methods.

Specific proposals could be made for the different analytical areas involved, so that EQAS reports on similar analytical areas would have the same structure. These proposals could include documents on how to design and manufacture EQAS reports, both the routine reports and the summary reports. They could also include other essential principals than those described in ISO Guide 43-1 (3) and ILAC–G13: 2000 (4) e.g. for EQAS operating over a certain period of time, the participants should receive historical data on individual performance from all samples sent out in that certain time period. Furthermore, the Internet could be used as a tool to improve the quality of reports in EQAS.

Harmonising the designs of EQAS reports, whether it would be a standardised single design or just harmonising distinct elements, can only evolve through international collaboration. It could be a task for the European Committee of External Quality Assurance Programme in Laboratory Medicine (EQALM) to set up an International Working Group, to establish and implement guidelines on designs and appropriate statistical evaluation procedures, modified to the individual analytical areas of EQAS routine and summary reports.

1. ISO/FDIS 15189. Quality Management in the Medical Laboratory. 2000.

2. ISO/IEC 17025. General Requirements for the Competence of Testing and Calibration Laboratories. 2000.

3. ISO/IEC Guide 43-1. Proficiency testing by interlaboratory comparisons – part 1: Development and operation of proficiency testing schemes. 2 ed. 1997.

4. ILAC-G13: 2000. Guidelines for the Requirements for the Competence of Providers of Proficiency Testing Schemes. 2000.

5. ISO 13528. Statistical methods for use in proficiency testing by interlaboratory comparisons. 2000.

In Cuba we have organized a network of laboratories to support several major screening programs associated with Epidemiological Surveillance, Blood Certification and Perinatal Care. All these programs are based on a common technical background: enzyme immunoassays that use only 10 µL of reagents and samples and a set of instruments that complement them, the Ultra Micro Analytic System (SUMA) produced by the Immunoassay Center (IAC).

At present around 110 laboratories are participating in an External Quality Assurance and Assessment Program (EQAAP) organized by IAC. The EQAAP carry out an integral assessment of the work of each participating laboratory (PL), detecting possible errors in pre-analytical and analytical steps, which allows us to define a directed service of technical attendance.

To offer an efficient technical assistance service to the laboratories network, the country was divided for practical reasons into 4 regions. A Regional Technical Assistance Center (RTAC), under the supervision of the IAC, was created in each of these regions.

In this paper we describe the software designed for data management and publication of the EQAAP.

Our software is formed from three subsystems. One of them installed on the PL, another on the RTAC and the last on the reference laboratory (RL).

The laboratory's subsystem has the responsibility for storage of all the work information in a period. This information is sent to the RTAC through electronic mail, which diminishes greatly the number of transcription errors. The RTAC’s subsystem stores the information received from each PL in their region and sends it to the RL. After the information sent by RTAC has been received by RL’s it is evaluated. The results of the evaluations are send to every RTAC and PL.

The system was developed using a database server to support the data of the RL’s subsystem. The server is connected to a computer network that permits access to all the available information. The PL and RTAC’s data are supported using a local database system. In the three subsystems the flow of information in both directions is carried out through electronic mail.

All the subsystems were developed using Borland Delphi 5.0 Enterprise Edition as the programming language. The PL and RTAC‘s local database systems are supported on Paradox 7.0 and the RL one on Microsoft SQL Server 6.5. The information is published through active web pages which query the database server and offer the results 100 % updated.

This system is being applied in our country since 1997 at the RL and RTAC. The PL’s subsystem is not completely installed yet due to some equipment requirements such as a modern computer, phone line and modem.

The application of this software has allowed the quality control of the work carried out in PL, the storage of the statistical information about all the health programs supported by the network of SUMA laboratories and the study of the performance of our assays post commercialization. We would like to emphasize that these studies are enriched with the information from the clinical confirmation.

This WHO 2^nd International Standard (IS) (code 98/612) is a valuable tool in the quality assurance of Plasma-Fibrinogen assays in Health laboratories. For exam-ple it is used for assigning values to the national coagulation cali-brator, for daily use in Danish routine laboratories provided by DEKS. It is also used in the establishment of reference values in Danish EQA.

In the routine laboratory Fibrinogen is determined in plasma from blood stabilised by addition of one volume of citrate solution to nine volumes of whole blood. In this way the original circulating plasma is diluted by about 15-20% depending of the original haematocrit. The amended "In-structions for Use" of the WHO 2^nd IS states that the assigned value is 2.2 mg Fibrinogen per ml after reconstitution with 1ml distilled water. The principle of standardisation of the WHO standard for fibrinogen is originally based on recording the mass of clotted fibrinogen in each ampoule gravimetrically. Therefore there is no sample dilution preconditioned in the stated value. Consequently for the assay of patients plasma one should correct the final result for the average of this dilution of the circulating patient plasma. This should be stated in the certificates of both the primary and secondary calibration materials.

According to several coagulation experts it is usual practice in coa-gulation laboratories to ignore this dilution of the circulating patient plasma, because they assume that the standardisation of this assay is done in the same manner as for e.g. INR, where the value assignment is based on the same type of plasma dilution. However this is not consolidated in the literature.

In many clinical biochemistry laboratories in e.g. Denmark and Sweden one USES the certificate value FOR CALIBRATION PURPOSE, IN THE METROLOGICAL WAY, and CORRECT the FINAL RESULT FOR the average sample dilution of patients blood.

However other laboratories neglect the dilution of the circulating plasma. These differences of daily practice are clearly observed in both the EQA results and the reported biological reference values ("normal values").

No individual is able to change laboratory habits. It is assumed that such change of practice in laboratories, needs to be recommended by international organisations before it can be generally accepted. The matter has therefore been brought forward to ISTH and the European Standardisation Organisation CEN, TC 140. It is assumed that a metrological example of establishing of the traceability of measurements of the substance concentration of Fibrinogen in plasma (circulating), will be incorporated into a forthcoming traceability standard.

While awaiting an international recommendation, a practical solution could be to use different nomenclature for the two types of quantities which originate from the different types of calibration. This could be done by referring to the different measurement system: Traditional calibration, Citrate plasma-Fibrinogen, …. ; and the "metrological" calibration, Plasma-Fibrinogen, …. This will probably also mean that different NPU-codes should be used when reporting results electronically. In countries where both the "metrological" and the traditional use of the WHO 2^nd IS for Fibrinogen are practised the relevant scientific national society may need to provide local recommendations.

Acknowledgement. The authors are indebted to Dr. Anthony R Hubbard, Principal Scientist, Division of Haematology, NIBSC, for good advice and helpful com-ments.

Third Millennium Standardisation in Specialised Laboratory Medicine

November 22 in Düsseldorf