Managing the risks of making the wrong diagnosis: First, do no harm

Abstract

The appropriate use of diagnostics is important as misdiagnosis may have serious consequences. Confidence in a diagnostic test result depends on the test's accuracy (sensitivity and specificity) in the context of the use-case (who is tested and why) and the prevalence of the condition investigated. Here, we offer an approach to diagnostics focused on the risks and effects of making the wrong diagnosis. We propose 'fitness brackets' for a given test to define the range within which the test is fit-for-purpose, based on the use-case and risk-management principles. We use as exemplars tests for dengue pre-vaccination screening and tests for diagnosing Covid-19 in different settings.

Keywords: Covid-19; Dengue; Diagnostics; Misdiagnosis.