Cholesterol measurements. Quality assurance and medical usefulness interrelationships.

This is a status report on cholesterol testing, using data from the College of American Pathologists Surveys program and Quality Assurance Service and the Lipid Research Clinics program. The clinical context for cholesterol testing is reviewed in order to develop a conceptual framework for analytic goals. Our analysis leads us to the following conclusions: First, the accuracy of cholesterol assays will be improved by using newly available reference material labeled by definitive methodology; this aspect of cholesterol testing is well monitored by proficiency testing. Tight goals for accuracy must await a consensus on the true value of test specimens. Second, long-term intralaboratory analytic variability, which is crucial in cholesterol testing, is well monitored by programs like the Quality Assurance Service, but not by proficiency testing. It is proposed that pathologists use realistic performance goals that incorporate a total error model and biologic variability.