Downstream outcomes: using insurance claims data to screen for errors in clinical laboratory testing.

Abstract

A methodology is described by which health insurance claims data might be used to discover the occurrence of systematic errors by clinical laboratories. False-positive results should generate a series of tests or treatments that are eventually abandoned as the false signal of the initial test is discovered while false-negative results may cause necessary tests or treatments to be unduly delayed. False results may also generate adverse outcomes such as an unusually high number of deaths or hospitalizations among persons who have received particular laboratory tests. Health insurance claims data may be used to discover these patterns and how the inclusion of laboratory results on claims would improve the precision of such inferences. Appropriate statistical tests are discussed.