The use of multiple-choice questions in medical assessments is ubiquitous and extends in breadth from undergraduate formative and summative assessments to postgraduate certifications and reaccreditations. Despite the clear importance of multiple-choice questions in medicine, in the last thirty years there has been no authoritative peer reviewed publication that reviews the overall generation and use of multiple-choice questions in medicine. Producing effective and useful medical multiple-choice questions is as much of an art form as it is a science - unfortunately, every new writer of such questions tends to rely on their own experiences of such questions and then travel along their individual and tortuous pathway to produce "acceptable" questions. Our review article provides practical guidance and information that will assist writers of medical multiple-choice questions. As well as extensively discussing common pitfalls, we provide a checklist that represents optimum practice in question generation. We directly compare examples of good and poor multiple-choice questions. We describe the metrics for measuring effective questions; specifically, the discrimination and difficulty indices. We discuss the optimum approach to training faculty to produce flawless multiple-choice questions. We describe the potential role and benefits of students in writing multiple-choice questions. We address the current and future effectiveness of the major artificial intelligence software brands in developing objectively functional multiple-choice questions. We continue to support the pragmatic use of MCQs in undergraduate and postgraduate medical training, with the caveat that ongoing training is made available to the creators in order to facilitate the production of the highest quality questions.
Steele et al. (Fri,) studied this question.