228 Chapter 7 to infection control measures, and forensic medicine and because of resources), and environment (to include sustainability)’. As an alternative, these aspects may also be considered when moving from evidence to decision (e.g. in considerations of acceptability and resources). Finally, the term ‘test’ is now referred to as ‘testing’, as this can include, for, example, a cascade of testing procedures, different frequencies of testing, and tests from different manufacturers. In terms of test performance, there has been a shift from diagnostic accuracy to test accuracy and clinical performance. Secondly, there has been a recurring debate regarding the definition of test burden. From an individual patient or consumer perspective (e.g. in the consultation room), burden can be considered as the practical demands that come with undergoing the process of a test that people may dislike, such as taking medication to prepare for a test or visiting the hospital [17]. In a broader sense, stress and costs associated with undergoing the test procedure can also be considered. From a population perspective, burden can also be defined as any undesirable aspect and consequence of testing, including adverse effects and complications related to the test. However, adverse effects and complications may also be considered people-important outcomes, possibly depending on their severity and prevalence. Agreeing clear definitions with all relevant stakeholders can solve any potential confusion. Thirdly, there is a noticeable discrepancy between the various purposes of testing and the evidence that has been published. The vast majority of literature on test evaluation and guideline development methods concerning testing focuses on diagnosing diseases [19, 21-23]. However, in healthcare practice and guideline development, it is also necessary to clarify the value of testing for other purposes. This includes among others prognostic and follow-up tests, such as a test to predict the likelihood of a particular event such as stroke, or a test to monitor the course of certain diseases, such as lung function in COPD. Additionally, it is possible for a single test to serve multiple purposes. For instance, mammography can be used as a screening instrument in women aged 50-75 years, as a diagnostic step in women with signs of breast cancer, and as a follow-up measure in women who have undergone breast cancer treatment. It is important for guideline developers to acknowledge the various possible purposes of testing. And fourth, when assessing the added value of tests, it is noteworthy that both test evaluation methods and guideline development methods, as well as this thesis, place great emphasis on measures of test accuracy, implying a dichotomy. However, it is important to note that test results can fall into different data types: they can be binary (e.g. a pregnancy test), ordinal (e.g. BI-RADS assessment categories ranging from 0 to 6), counts (e.g. complete blood count), or continuous (e.g. body temperature). Furthermore, it should be realised that in healthcare practice, a test is usually
RkJQdWJsaXNoZXIy MTk4NDMw