Mariska Tuut

166 Chapter 4 In the GRADE Guidelines Series, Schünemann et al. presented evidence-to-decision frameworks for tests in clinical practice and public health [6]. In this guidance, attention is given to aspects that typically belong to the evaluation of medical tests: - ‘Formulating a question about a test in the PICO format should include the test and subsequent management strategies as pathways to important outcomes as well as identifying subgroups that might require different recommendations or options.’ This is also known as the test-treatment pathway. - Determination whether the problem is a priority: this depends on the perspective (e.g., individual patient or population). For screening topics, this could also be a public health perspective. - ‘Judgments about test accuracy should be based on a summary of findings from a systematic review of test accuracy studies’. - Benefits and harms: ‘Judgments about the benefits and harms of using a test require preparation of a summary of findings for the modelled desirable and undesirable effects on health outcomes’. And: ‘This includes information about direct benefits and harms of the test and the downstream consequences of interventions. In particular, judgments about the effects of the interventions that follow based on the test results (linked evidence) should be informed by a summary of findings table’. - ‘Rating the certainty of the evidence for the effects of tests requires consideration of each element of the linked evidence used to inform judgments about their benefits and harms’. This includes certainty of the evidence of:  Test accuracy.  Test related direct benefit, adverse effect, or burden of the test.  Natural course of the condition and the effect of management guided by test results.  Link between test result and management decisions.  Overall quality of the evidence. - Valuing main outcomes: ‘For tests, this includes adverse effects and any burden associated with the test, as well as downstream outcomes of linked interventions’. - Balance between desirable and undesirable effects: ‘For tests, this judgment is informed by the results of either formal or informal modelling of the anticipated desirable and undesirable effects of linked interventions’. - Resource use: ‘This includes judgments about how large the resource requirement was, the certainty of the evidence of resource requirement and the costeffectiveness of interventions. This includes consideration of downstream costs….’.

RkJQdWJsaXNoZXIy MTk4NDMw