Basic statistics for quality control in the clinical laboratory.

Abstract

A systematic approach toward selection and evaluation of laboratory methods should improve the quality of the analytical service. However, any testing approach only assesses whether the analytical method can achieve the desired level of performance under the test conditions. It does not assure that the performance is actually achieved under the conditions of each routine analytical run. The quality of the routine analysis must be monitored by suitable quality control procedures. Unfortunately, the present practice of quality control achieves an arbitrary and usually unknown level of quality. Achieving a known level of quality requires that one understand the performance characteristics of the quality control procedures themselves. Achieving a specified level of quality requires that one design the quality control system to achieve a stated error specification. This can be accomplished by careful choice of control procedures in order to have the desired capability for detecting analytical errors which would invalidate the medical usefulness of the test results. A high-caliber quality control system will be the result of a logical sequence of the objective evaluation and selection of an optimal analytical method followed by regular rigorous statistical monitoring of each analytical method.