Descriptive Analysis of the Psychometric Properties of Extended Matching Questions Conducted Among Anaesthesia Residents

Background: Clinical reasoning is one of the core features of clinical competency. Training and assessing clinical reasoning is vital in postgraduate training. Extended matching questions􀀁(EMQs) are effective in assessing problem-solving and clinical reasoning abilities, but not commonly used in Postgraduate training. Covid-19 pandemic, which prevented both patient encounter and regular academic activities, warranted the introduction of innovative Teaching-Learning methods to sustain clinical reasoning skills. Hence, we aimed to introduce EMQs in formative assessment among anesthesia residents and analyzed its psychometric properties. Methods: The study was conducted at the Department of Anaesthesiology, Mahatma Gandhi Medical College and Research Institute (MGMCRI), Pondicherry, India. Four modules of EMQs as part of a formative assessment was conducted among residents (n=20). A total of 40 clinical vignettes and 60 options were administered. Post-validation of the EMQs was done by item analysis. Test reliability was estimated by the Kuder-Richardson 20 formula. Difficulty index (DIF-I), discrimination index (DIS-I), and distractor functionality were analyzed. Results: The KR-20 reliability coefficient was 0.72. The mean DIF-I was 0.43±0.17, from which 72.5% (29) were in the acceptable range, 20% (8) difficult, and 7.5% (3) easy. The mean DIS-I was 0.28±0.24, where 40% (16) had acceptable, 27.5% (11) excellent, and 20% had poor discrimination. Ninety percent of distractors were functional. The DISI exhibited a positive correlation with DIF-I (r= 0.2155, P=0.0185). Conclusions: The results of the present study indicated that EMQs have acceptable test reliability. The majority of the items (80%) followed the principles of MCQs. We concluded that EMQs can be effectively used as part of the postgraduate assessment to test higher-order knowledge and clinical competency