Inter-laboratory quality control with investigation of different methodological characteristics

Abstract

A classification system was designed which gives a brief description of the analytical methods used by laboratories participating in an external quality control program. Together with the analytical results several characteristics of the methods are to be reported by the laboratories, e.g. principle, special variant, commercial reagent, type of calibration and of instrumental equipment. The system is logically structured to be easy to learn and to facilitate the transfer of the information to a computer. To test this system, a survey was made with forty-two participating large or middle sized Nordic laboratories. Control material at two concentration levels was analysed on ten different occasions. A computer program was used which allowed statistical treatment of the data and grouping of the results according to different outlined characteristics of the method. For some methods, but not all, the use of calibration standards, prepared with weighed amounts of the components were superior to calibrations using commercial sera with assigned concentration values. Automated analysis seemed to give higher analytical quality than manual with the exception of analysis of S-bilirubin. The results furthermore indicated that the analytical quality was related to type of equipment and that three commercial enzyme reagents from one company gave higher within- and between laboratory variations than the corresponding reagents from another company. The general analytical performance including intra- and inter-laboratory variation, was similar to that reported from other surveys. Except for tests of significance F-values were calculated in order to compare between- and within laboratory variations. It is suggested that methods with high. F-values be investigated with regard to their calibration procedures.

Key Words:

• analytical methods
• characterization
• inter-laboratory
• intra-laboratory variation
• quality control