Properties and units in the clinical laboratory sciences. V. Properties and units in thrombosis and haemostasis: ISTH-IUPAC-IFCC recommendations 1995.

For historical reasons, the terms used in the nomenclature for properties in thrombosis and haemostasis differ according to 'school' of thought. This hampers communication. In collaboration, The Scientific and Standardization Committee of the International Society on Thrombosis and Haemostasis and the Committee (Commission) of Quantities and Units (in Clinical Chemistry) have prepared a set of recommended systematic names for properties in that domain. For use in electronic transmission each property has been given a code value.

Reduction of analytical error in measurement of the concentration of coagulation factors in plasma.

A modification of the method of Owren & Aas for the determination of relative arbitrary substance concentration of coagulation factors (II+VII+X) in plasma is presented. The coefficient of variation on repetends (today-yesterday) is reduced to 0.03 for results in the therapeutic as well as the 'normal' interval. An unusual composition of the plasma may be revealed. Thromboplastin-absorbed bovine plasma should be kept at -70 °C.

Mathematical expressions of measuring curves for concentration of clotting factors.

Conversion of coagulation time (t) to relative arbitrary substance concentration of coagulation factors (II + VII + X) abbreviated c(ra)) was originally made from a measuring curve corresponding to the equation log t = a(o) +a1 log c(ra). Three more suitable equations are presented and tested for fit: t = A(o) + A1 c(ra); t = B(o) + B1 c(ra) + B2 c²ra; log t = b(o) + b1 log c(ra) + b2 log c(ra)². Their suitability for the Simplastin®-A and Thrombotest® methods is also investigated. The last equation yields the best fit and replaces the error-prone reading of log-log curves by a calculation suitable for routine use.

Elements of good practice in decentralized clinical laboratories.

The possibilities of producing decentralized clinical laboratory data 'nearer the patient' has augmented rapidly during the last decennium due to both simple and sophisticated equipment, often intended to be operated by nonlaboratorians. The theory and practice of quality assurance in its wider sense has not kept pace with this development. The components of Good Laboratory Practice are presented under the headings: type of laboratory work, discipline, management, personnel, premises, safety, equipment, reagents, standard operating procedures, internal quality control, external quality assessment, method, dedicated operating procedure, syllabi, and clinical relevance. The projects currently being established by different national, regional, and international bodies to formulate guidelines should be coordinated to avoid duplication and conflict.

Scales for measurement based on an antigen-antibody reaction.

In spite of sometimes complicated reaction sequences, the multitude of immunomethods all rely on the principle of a reaction between an antigen and an antibody and follows physico-chemical kinetics according to molecular relationships. The preferred kind-of-quantities for amount of analyte, therefore, should be number of entities (unit: 1) and amount-of-substance (mol)--both requiring definition of the elementary entity of the analyte--rather than mass (kg) and arbitrary amount-of-substance(procedure) (arbitrary unit). Derived kind-of-quantities for concentration (dividing by volume of system) and content (dividing by mass of system) are easily defined. The measurement scale is an ordered set of possible values that the quantity may take. Any scale should be described as to type and the magnitudes and number of possible values. The types comprise nominal (no magnitudes), ordinal (inequality of magnitudes), difference (equality of differences), and ratio scales (equality of ratios). Each scale is characterized by the statistics allowed with the measured values. The magnitudes of possible values are derived from the nature of the quantities being considered and the measurement procedure. The number of possible values will vary from two to many and depends on the uncertainty of results delivered by the procedure. With this systematic description, the ambiguous terms "qualitative" and "semiquantitative" may be abandoned. A given analyte in a certain type of system may give rise to values on any of the scales if it can be measured on a ratio scale because simpler and chemically less informative results may be obtained by transformation.

General metrological requirements of clinical lipid measurements.

Quantities in blood, plasma, or serum having lipid components are often difficult to define, especially as regards the component which must be selected according to chemical, biological, and clinical considerations. The choice of generic quantities and their respective measurement procedures relates to clinical requirements of allowable uncertainty, effectiveness, comparability, and compatibility determining goals of analytical reliability. The measurement procedure should define preanalytical requirements and be based upon traceability from tertiary and secondary reference materials with reference procedure values to primary reference materials. Such materials and procedures should be established internationally. A comprehensive quality assurance system of internal quality control and external quality assessment is essential to ensure worldwide and continued comparability. Validation and transferability should also be demonstrated by regional or international collaboration. Transmutation of results between procedures requires extensive investigations.

Measurement, value, and scale.

The terms 'qualitative', 'semiquantitative', and 'quantitative' are used ambiguously. Based on international recommendations by IEC, IFCC, ISO, IUPAC, and OIML, and the work on scales by Stevens, a systematic terminology is presented. Measurement is considered to be the set of operations by which a value (consisting of a relational operator, and symbols, figures, or letters) is assigned to a quantity. The possible values constitute a scale that may be subdivided into classes. A hierarchy of four types of scale is characterized: nominal (values are independent of magnitude), ordinal (ranked according to magnitude), interval (equality of differences, arbitrary zero), and ratio scale (equality of ratios, absolute zero); each type allows a different set of statistical calculations. The type of scale is independent of its number of values or classes, or the uncertainty of measurement. The number of values on a scale is indicated by the terms two-value, three-value,..., multivalue and the number of classes analogously by two-class, etc. Examples of transformation of quantities are given.

A quality manual for the clinical laboratory including the elements of a quality system. Proposed guidelines.

Development of quality manuals is a means for the promotion of quality in clinical laboratories by describing the total quality system. It also provides opportunity of checking whether the quality system is implemented in reality and demonstrates to the hospital administration and the clinicians that the laboratory is committed to quality. The intention of these guidelines is to describe the elements of the quality system for a large clinical laboratory, and to present such a system in the form of a quality manual. The proposed guidelines comply, where relevant, with ISO/IEC guide 25 'General requirements for the technical competence of testing laboratories' and EN 45001 'General criteria for the operation of testing laboratories'. The document may be used as an aid for laboratories wishing to be accredited according to EN 45001, or intending to apply for formal certification of their quality systems, according to ISO 9001 'Quality systems--Model for quality assurance in design/development, production, installation, and servicing' utilizing ISO 9004 'Quality management and quality system elements--guidelines; Part 2 Guidelines for service'. However, information about the minimum requirements for official recognition should be obtained from the particular accreditation or certification body concerned.