Dong Ting, Durning Steven J, Artino Anthony R, van der Vleuten Cees, Holmboe Eric, Lipner Rebecca, Schuwirth Lambert
Department of Medicine, Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD 20814.
Maastricht University, P.O. Box 616, 6200 MD Maastricht, The Netherlands.
Mil Med. 2015 Apr;180(4 Suppl):92-6. doi: 10.7205/MILMED-D-14-00555.
Clinical reasoning is essential for the practice of medicine. Dual process theory conceptualizes reasoning as falling into two general categories: nonanalytic reasoning (pattern recognition) and analytic reasoning (active comparing and contrasting of alternatives). The debate continues regarding how expert performance develops and how individuals make the best use of analytic and nonanalytic processes. Several investigators have identified the unexpected finding that intermediates tend to perform better on licensing examination items than experts, which has been termed the "intermediate effect."
We explored differences between faculty and residents on multiple-choice questions (MCQs) using dual process measures (both reading and answering times) to inform this ongoing debate.
Faculty (board-certified internists; experts) and residents (internal medicine interns; intermediates) answered live licensing examination MCQs (U.S. Medical Licensing Examination Step 2 Clinical Knowledge and American Board of Internal Medicine Certifying Examination) while being timed. We conducted repeated analysis of variance to compare the 2 groups on average reading time, answering time, and accuracy on various types of items.
Faculty and residents did not differ significantly in reading time [F (1,35) = 0.01, p = 0.93], answering time [F (1,35) = 0.60, p = 0.44], or accuracy [F (1,35) = 0.24, p = 0.63] regardless of easy or hard items.
Dual process theory was not evidenced in this study. However, this lack of difference between faculty and residents may have been affected by the small sample size of participants and MCQs may not reflect how physicians made decisions in actual practice setting.
临床推理是医学实践的关键。双重加工理论将推理概念化为两大类：非分析性推理（模式识别）和分析性推理（对备选方案进行积极比较和对比）。关于专家表现如何发展以及个体如何最佳利用分析性和非分析性过程的争论仍在继续。几位研究者发现了一个意外的现象，即中级水平者在执照考试项目上的表现往往比专家更好，这一现象被称为“中级效应”。
我们使用双重加工测量方法（阅读时间和答题时间）来探究教员和住院医师在多项选择题上的差异，为这一持续的争论提供信息。
教员（获得委员会认证的内科医生；专家）和住院医师（内科实习医生；中级水平者）在计时的情况下回答实际执照考试的多项选择题（美国医师执照考试第二步临床知识考试和美国内科医学委员会认证考试）。我们进行了重复方差分析，以比较两组在各类题目上的平均阅读时间、答题时间和准确率。
无论题目难易，教员和住院医师在阅读时间[F(1,35)=0.01,p=0.93]、答题时间[F(1,35)=0.60,p=0.44]或准确率[F(1,35)=0.24,p=0.63]方面均无显著差异。
本研究未证实双重加工理论。然而，教员和住院医师之间缺乏差异可能受到了参与者样本量小的影响，而且多项选择题可能无法反映医生在实际临床环境中如何做出决策。
