It is important to avoid adopting the ‘default’ approach with respect to fundamental processes such as item development. However, this is not always achieved. For example: despite a strong body of research spanning 30+ years suggesting that three-option (one keyed option and two distractors) multiple-choice items are optimal, it is estimated that only about 1% of contemporary high-stakes assessments contain such items (Edwards, Arthur & Bruce, 2012). Indeed, the majority of test developers across a wide range of contexts and disciplines favour the use of 4- or 5-option items, with no justification for this practice. This session will provide an overview of practical, fiscal and psychometric arguments supporting the use of 3-option items. Furthermore, it will describe and discuss the findings of an empirical research study comparing stem-equivalent 3- and 4-option items, in the context of a food safety certification test administered to over 1,000 candidates.