“Pass, fail” – On Standard Setting Procedures for the Assessment of Practical Skills at Medical Schools in Germany, Austria, and Switzerland

Themembers of the GMA committee on assessment have the common goal to make exams in the training and continuing education of health professionals in Germany, Austria and Switzerland (D/A/CH countries) as appropriate as possible. For this purpose, they engage in a discourse with each other, considering both scientific research and regional/local contexts. The committee is pleased to present in the present special issue its views on practical skills from the assessment perspective. The teaching of practical skills is now well-established in the medical programs in the D/A/CH region, reflected in the national catalogues of learning objectives: The Swiss Catalogue of Learning Objectives for Undergraduate Medical Training (SCLO) [1], the Austrian Competence Catalogue for Medical Skills (ÖKÄF) [2], and the German National Competence Based Catalogues of Learning Objectives for Undergraduate Medical Education (NKLM) [3]. The assessment of practical skills, too, is largely established. As early as 2007, over 70% of Germany’s medical programs featured practical skills assessment, most of which in OSCE format [4], and since 2011, an OSCE complements the Swiss Federal Licensing Examination andmust be passed independently of the MCQ examination [5]. Besides OSCE, the oral-practical examination [6] is a common format, but also OSPEs (Objective Structured Practical/Preclinical Exams) are used [7]. This shows that practical examinations in medical education are indeed varied and widespread, but what about their quality? Yudkowsky lists various criteria compromising construct validity – i.e., the meaningful interpretability of the data obtained in an exam [8]. As is the case with exams of other formats, it is necessary in the assessment of practical skills such as the OSCE to prevent construct underrepresentation by providing sufficient and representative examination content and tominimize construct-irrelevant variance, i.e., to ignore, in interpreting exam performance, such information not relating to the subject matter under scrutiny. Quality-maintaining measures are being taken in many programs, such as blueprinting (selection of adequate, representative exam content), review (e.g., with respect to wording of tasks and checklists, and to anticipate task difficulty), as well as training of examiners and possibly, simulated patients involved. However, it seems that one aspect of quality assurance has been receiving only little attention: the deliberate