Kenneth Lippel

Automated determination of total plasma cholesterol: a serum calibration technique

The automated (AutoAnalyzer II) determination of cholesterol in nine serum pools in the concentration range 3.465-8.871 mmol/l, gave results that were approximately 10% higher than reference values when the analyses were based on unesterified cholesterol standards containing the same amount of water as the sample extracts (1963 analyses in 12 laboratories during a 12 month period; automated value = 0.032 + 1.10 X (Reference value)). A serum calibration procedure was successful ilues, and was equally effective in correcting the values observed for aliquots of 368 fresh-frozen plasma samples analyzed in each of the 12 laboratories during a 38 month period.

Central laboratory quality control in the National Cooperative Gallstone Study.

This article presents the rationale, selection, operation, and quality control of the Central Serum and Central Bile Laboratories utilized by the National Cooperative Gallstone Study. The external quality control protocols were designed to monitor long-term stability of the analytical procedures and to measure the precision of the measurements as affected by the collection, labelling, storage, shipping, and laboratory methods. For both laboratories, the assessment of long-term stability by pool standards failed to produce the data necessary to come to relevant conclusions. Several of the problems involved, however, did lead to protocol changes that increased the reliability of the laboratory data. The use of duplicate measurements to monitor precision was more successful and demonstrated acceptable performance of these systems. This article describes the external quality control surveillance procedures employed in the NCGS, their strengths and weaknesses, statistical methods for the analysis of such quality control programs, and the implications for the final statistical analyses of the clinical trial patient data.