Back

The Schema Method: Using Quantified Learning to Innovate, Augment, Assess and Analyze Learning in Medicine

Sebin, D.; Doda, V.; Balamani, S.

2022-09-07 medical education
10.1101/2022.09.06.22279624 medRxiv
Show abstract

IntroductionThe predominant method of learning Medicine at its core has remained unchanged for decades. This stagnancy creates a need for making learning more effective, insightful, and quantified. Schema achieves this through e-learning, active feedback, and quantified learning by granulating the medical curriculum into specific subtopics selected based on the crucial knowledge that a competent medical learner must possess, hereafter referred to as yield. This particular study aims to analyze medical students multidimensional competency in solving clinical scenario-based MCQs pertaining to vertically integrated topics derived from the Schema. MethodsA retrospective study was conducted by analyzing the user data of a leading e-learning platform for medical students. For the purposes of this study, twenty such "high-yield" Schema topics were shortlisted as being the most crucial knowledge areas. Students performance in solving a fixed set of SOC-MCQs of these Schema topics was used to gauge their competence. Performance variation over five years (2018-2022) was also analyzed to study the changing patterns in topic-specific performance. ResultsA total of 20 Schema topics were shortlisted, consisting of 128 MCQs. The number of participants solving each Single Option Correct Multiple Choice Question (SOC-MCQ) ranged from 60,080 to 2,06,672. A significant improvement in the Net Delta was observed in 9 topics. Performance in other topics showed either no significant change or a significant downtrend. ConclusionA significant performance uptrend (ND = 128%) was observed in Anaphylaxis, Basic Lifesaving Skills, ST-Elevated Myocardial Infarction, Glasgow Coma Scale, and subdural hemorrhage & Syndromic management of Sexually Transmitted Infections, most of which are either acute or emergency conditions. A significant downtrend in performance was seen in Schema topics such as Asthma management, Hypertension management, Diabetic Ketoacidosis, and Subarachnoid hemorrhage pertaining to chronic conditions. Several hypotheses for these findings can be derived, the validities and collective impacts of which can be explored in more in-depth and broader studies in the future.

Matching journals

The top 1 journal accounts for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.