Carey-Ewend Kelly, Feinberg Amir, Flen Alexis, Williamson Clark, Gutierrez Carmen, Cykert Samuel, Beck Dallaghan Gary L, Gilliland Kurt O
Gillings School of Global Public Health, Chapel Hill, NC USA.
University of North Carolina at Chapel Hill School of Medicine, Chapel Hill, NC USA.
Med Sci Educ. 2023 Apr 12;33(3):659-667. doi: 10.1007/s40670-023-01778-z. eCollection 2023 Jun.
This paper aims to characterize the use of demographic data in multiple-choice questions from a commercial preclinical question bank and determine if there is appropriate use of different distractors.
Multiple-choice questions for medical students often include vignettes describing a patient's presentation to help guide students to a diagnosis, but overall patterns of usage between different types of nonmedical patient information in question stems have yet to be determined.
Three hundred eighty of 453 randomly selected questions were included for analysis after determining they contained a clinical vignette and required a diagnosis. The vignettes and following explanations were then examined for the presence/absence of 11 types of demographic information, including age, sex/gender, and socioeconomic status. We compared both the usage frequency and relevance between the 11 information types.
Most information types were present in less than 10% of clinical vignettes, but age and sex/gender were present in over 95% of question stems. Over 50% of questions included irrelevant information about age and sex/gender, but 75% of questions did not include any irrelevant information of other types. Patient weight and environmental exposures were significantly more likely to be relevant than age or sex/gender.
Students using the questions in this study will frequently gain practice incorporating age and sex/gender into their clinical reasoning while receiving little exposure to other demographic information. Based on our findings, we posit that questions could include more irrelevant information, outside age and sex/gender, to better approximate real clinical scenarios and ensure students do not overvalue certain demographic data.
The online version contains supplementary material available at 10.1007/s40670-023-01778-z.
本文旨在描述商业临床前题库中多项选择题对人口统计学数据的使用情况，并确定是否合理使用了不同的干扰项。
医学生的多项选择题通常包括描述患者表现的病例 vignette，以帮助引导学生做出诊断，但题干中不同类型非医学患者信息的总体使用模式尚未确定。
在确定453个随机选择的问题包含临床 vignette 并需要做出诊断后，纳入其中380个问题进行分析。然后检查 vignette 及其后的解释中是否存在11种人口统计学信息，包括年龄、性别和社会经济地位。我们比较了11种信息类型之间的使用频率和相关性。
大多数信息类型出现在不到10%的临床 vignette 中，但年龄和性别出现在超过95%的题干中。超过50%的问题包含与年龄和性别无关的信息，但75%的问题不包含其他类型的无关信息。患者体重和环境暴露比年龄或性别更有可能是相关的。
使用本研究中问题的学生在将年龄和性别纳入临床推理时会经常得到练习，而接触其他人口统计学信息的机会很少。基于我们的发现，我们认为问题可以包括更多年龄和性别之外的无关信息，以更好地模拟真实临床场景，并确保学生不会过度重视某些人口统计学数据。
在线版本包含可在10.1007/s40670-023-01778-z获取的补充材料。
