Item Analysis of Otorhinolaryngology (ENT) Multiple-Choice Questions in an Internal Assessment Among Undergraduate Medical Students at a Rural Medical College in Maharashtra, India: An Analytical Cross-Sectional Study

Background: Multiple-choice questions (MCQs) are widely used for the assessment of knowledge in undergraduate internal assessment, but systematic post-test evaluation is required to maintain the quality of MCQ items. This study aimed to evaluate an otorhinolaryngology (ENT) internal assessment MCQ paper using difficulty index, discrimination index, and distractor efficiency to identify items for retention, revision, or discard to strengthen the departmental MCQ bank. Methods: An analytical cross-sectional item analysis was conducted for the seventh-semester MBBS ENT Part Completion Test at JIIU's Indian Institute of Medical Sciences and Research Medical College, a rural medical college in Maharashtra, India. All students who appeared for the exam were included (n = 138). The test comprised 30 single best-answer MCQs (one key and three distractors). There was no negative marking. Difficulty index (p-value), discrimination index (DI, extreme-group method; upper and lower 27%), and distractor efficiency (DE; non-functional distractor defined as <5% selection) were calculated. Items were categorized as retain/revise/discard based on prespecified rules. Results: The mean total score was 10.76 ± 5.36 (range 4-28). The mean difficulty index was 0.359 (range 0.072-0.833); 15/30 (50%) items were difficult (p < 0.30), 14/30 (46.7%) were of moderate difficulty (0.30-0.70), and 1/30 (3.3%) was easy. The mean discrimination index was 0.425; 16/30 (53.3%) items showed good discrimination (DI ≥ 0.40), 8/30 (26.7%) acceptable (0.20-0.39), and 6/30 (DI < 0.20) poor discrimination, including three items with negative discrimination (Q6, Q7, and Q25). The mean distractor efficiency was 90.0%; 23/30 (76.7%) items had DE = 100%, 5/30 (16.7%) had DE = 66.7%, and 2/30 (6.7%) had DE = 33.3%. Five items (5/30; 16.7%) met the predefined desired criteria (moderate difficulty, DI ≥ 0.40, DE = 100%). Overall, 10 items were retained, 17 recommended for revision, and three discarded. Conclusion: The ENT internal assessment contained many items with good discrimination and strong distractor efficiency, but the difficulty index was skewed toward difficult items. Three items showed negative discrimination. Routine item analysis with structured revision can strengthen the departmental MCQ bank and support local assessment quality improvement.