General introduction 19 1 specificity: the probability of getting a negative test result in people without the target condition (TN/(TN+FP)) positive predictive value (PPV): the probability of having the target condition in people with a positive test result (TP/(TP+FP)) negative predictive value (NPV): the probability of not having the target condition in people with a negative test result (TN/(TN+FN)) Clinical effectiveness (also known as clinical utility): this refers to the ability of a test to improve peopleimportant outcomes. Cost-effectiveness: this refers to the assessment of changes in costs and people relevant outcomes resulting from the introduction of a test. There are several perspectives from which costs can be determined, such as the individual patient perspective (e.g. costs that patients have to pay to undergo a test), the healthcare perspective (e.g. costs because of time invested by healthcare professionals and other resources needed for performing a test) and societal perspective (e.g. costs of testing covered by health insurance). In all perspectives, direct costs (such as costs of the tests), and indirect costs can be taken into account. Indirect costs could include travel expenses and costs for childcare for the patient while travelling to the hospital, loss of income, and social security expenses due to absence at work. Broader impact: this refers to consequences of the test beyond clinical effectiveness and costeffectiveness, such as acceptability, implementability, and consequences on legal, ethical, and organisational issues. Besides the above-mentioned evaluation scenarios, it is essential to define the role of a new test in comparison to the existing test, as this influences the interpretation of the new test’s value. Various roles are acknowledged [47, 48]: - Triage - Replacement of a reference standard or an existing test - Add-on - Parallel/combined These roles are described in detail in table 2. All of the above factors can be relevant when considering the benefits and harms of testing in specific circumstances and for specific populations. Impact of inappropriate testing There is considerable practice variation in test usage in practice, with both underuse and overuse of tests being common [49, 50]. Sullivan et al. conducted a systematic review on over- and undertesting in primary care, in which they explored the frequency of inappropriate ordering of 103 diagnostic tests in relation to their respective guidelines. The results showed a wide range of non-compliance to the testing recommendations in guidelines (median: 40.0%; range: 0.2-100%). Examples of underuse (inappropriately not performed tests) include echocardiography for heart failure (89% underuse) or atrial fibrillation (56% underuse), and pulmonary function tests for COPD (73% underuse). Examples of overuse (inaccurately performed tests) include echocardiography in people with no symptoms or signs of cardiovascular disease (77-92% overuse), urine cultures (77% overuse), upper gastrointestinal
RkJQdWJsaXNoZXIy MTk4NDMw