What information on quality specifications should be communicated to clinicians, and how?

Abstract

Background: Quality specifications, the level of performance required to facilitate clinical decision-making, not only have a central role in quality management in the laboratory but are also essential for assuring the interpretation and utilization of laboratory data by physicians. Consensus has been reached on the hierarchy of criteria for quality specifications. However, the information on quality specifications that should be communicated to clinicians, and the way in which this information should be given, is still widely debated. Methods: Laboratory tests have been grouped into four categories including uni-modal, and bi-modal distributions, tests used in patients monitoring and in evaluating the response to therapy (serial measurements), and, finally, tests that require interpretive comments. The most suitable and informative ways to communicate quality specifications to clinicians have been proposed for each category. Results: For tests with a uni-modal distribution, the decision limits should replace traditional reference values in the report. For tests with a bi-modal distribution, in addition to traditional reference values, some flags based on the uncertainty (i.e., analytical and biological variability) of laboratory data, can be included to help clinicians interpret laboratory data. For tests used in monitoring patients and in evaluating the response to therapy (serial measurements), the reference change value or the most effective threshold of the difference between two consecutive results should be indicated. For tests/test batteries that require interpretive comments, these should be added to the report and discussed in multidisciplinary meetings and interpretive rounds to promote knowledge of a more objective evaluation of laboratory data. Conclusions: A proposal has been made to improve the way laboratory results are communicated to clinicians, with practical information derived from quality specifications. By providing clinicians with information on quality characteristics and the degree of uncertainty, a more objective interpretation of laboratory data may be possible, and data may be more appropriately utilized for diagnosis and monitoring.