Medical Education
Conditions
Keywords
Artificial intelligence, Medical education, BOPPPS, Case-based learning, Clinical reasoning, Nephrology clerkship
Brief summary
This prospective, three-arm, parallel-group, cluster randomized controlled educational trial was conducted among third-year undergraduate medical students undertaking nephrology clerkship at The Second Hospital of Shandong University. The study aimed to compare the effects of lecture-based learning, BOPPPS combined with case-based learning, and artificial intelligence-integrated BOPPPS combined with case-based learning on students' theoretical knowledge, clinical reasoning, clinical case analysis, and critical appraisal of AI-generated clinical recommendations. Existing clinical clerkship teaching groups, rather than individual students, were randomized to one of three educational interventions: lecture-based learning (LBL), BOPPPS combined with case-based learning (BOPPPS-CBL), or artificial intelligence-integrated BOPPPS combined with case-based learning (AI-BOPPPS-CBL). Twelve teaching groups were randomized in a 1:1:1 ratio, with four teaching groups assigned to each arm. All groups received standardized instruction on acute kidney injury and chronic renal failure delivered by the same teaching team. The BOPPPS-CBL intervention incorporated structured learning activities including case discussion, progressive case disclosure, clinical reasoning exercises, and feedback. The AI-integrated group additionally used AI-generated clinical scenarios and diagnostic and treatment recommendations as materials for critical appraisal. Students identified potential errors or limitations in AI-generated content and proposed revisions based on textbooks, clinical guidelines, and evidence-based resources. The primary outcome was the clinical case analysis score. Secondary outcomes included theoretical test scores, knowledge gain, clinical reasoning score, AI critical appraisal score, and course feedback.
Detailed description
This prospective, three-arm, parallel-group, cluster randomized controlled educational trial was conducted among third-year undergraduate medical students undertaking nephrology clerkship at The Second Hospital of Shandong University. Existing clinical clerkship teaching groups served as the units of randomization. Twelve teaching groups were randomized in a 1:1:1 ratio to lecture-based learning (LBL), BOPPPS combined with case-based learning (BOPPPS-CBL), or artificial intelligence-integrated BOPPPS combined with case-based learning (AI-BOPPPS-CBL), with four teaching groups assigned to each arm. All groups received standardized instruction on acute kidney injury and chronic renal failure delivered by the same teaching team. The BOPPPS-CBL intervention incorporated structured learning activities including case discussion, progressive case disclosure, clinical reasoning exercises, and feedback. The AI-integrated group additionally used AI-generated clinical scenarios and diagnostic and treatment recommendations as materials for critical appraisal. Students identified potential errors or limitations in AI-generated content and proposed revisions based on textbooks, clinical guidelines, and evidence-based resources. The primary outcome was the clinical case analysis score. Secondary outcomes included theoretical test scores, knowledge gain, clinical reasoning score, AI critical appraisal score, and course feedback.
Interventions
BEHAVIORALLecture-Based Learning
Students received traditional lecture-based learning during nephrology clerkship education. Standardized teaching on acute kidney injury and chronic renal failure was delivered by the teaching team using conventional instructional methods.
BEHAVIORALBOPPPS Combined With Case-Based Learning
Students received a structured BOPPPS teaching model combined with case-based learning activities during nephrology clerkship education. The intervention included case discussion, progressive case disclosure, clinical reasoning exercises, and feedback from instructors.
BEHAVIORALAI-Integrated BOPPPS Combined With Case-Based Learning
Students received an artificial intelligence-integrated BOPPPS case-based learning model. AI-generated clinical scenarios and diagnostic and treatment recommendations were used as learning materials for critical appraisal. Students evaluated AI-generated content and identified potential errors or limitations based on clinical guidelines and evidence-based resources.
Sponsors
The Second Hospital of Shandong University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
OTHER
Masking
SINGLE (Outcomes Assessor)
Masking description
Students and instructors could not be blinded because of the nature of the educational interventions. Examination papers were coded without group information, and personnel responsible for scoring the written assessments were blinded to group allocation. Personnel performing the statistical analysis were also blinded to group allocation during the primary analysis.
Intervention model description
This was a three-arm, parallel-group, cluster randomized controlled educational trial. Existing clinical clerkship teaching groups were the units of randomization. Twelve teaching groups were randomized in a 1:1:1 ratio, with four teaching groups assigned to each arm. Students within each teaching group received the same assigned educational intervention.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Third-year undergraduate medical students in the clinical medicine program of Shandong University undertaking the nephrology clerkship at The Second Hospital of Shandong University. * Completion of the required theoretical courses in pathophysiology and internal medicine before study participation. * Provision of written informed consent.
Exclusion criteria
\- Previous nephrology clinical clerkship or internship experience, or previous participation in additional nephrology-focused training beyond the standard pre-clerkship curriculum.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Study-Specific Clinical Case Analysis Assessment Total Score | Immediately after completion of the educational intervention | Clinical case analysis performance was assessed using a study-specific three-case written assessment developed by the nephrology teaching team. The assessment evaluated students' ability to apply nephrology knowledge to clinical case analysis, diagnostic reasoning, treatment planning, and critical appraisal of AI-generated clinical recommendations. Total scores ranged from 0 to 70 points, with higher scores indicating better overall clinical case analysis performance. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Study-Specific Post-Class Theoretical Knowledge Test Score | Immediately after completion of the educational intervention | Theoretical knowledge was assessed using a study-specific post-class test covering acute kidney injury and chronic kidney disease. The test consisted of 15 single-choice questions, with each correct answer worth 2 points. Total scores ranged from 0 to 30 points, with higher scores indicating better theoretical knowledge performance. |
| Study-Specific Clinical Case Analysis Assessment: Clinical Reasoning Component Score | Immediately after completion of the educational intervention | Clinical reasoning was assessed using Cases 1 and 2 of the study-specific Clinical Case Analysis Assessment. The cases addressed acute kidney injury and chronic kidney disease and evaluated students' ability to identify clinical problems, interpret clinical and laboratory information, formulate diagnoses, assess complications, and develop appropriate management plans. Scores ranged from 0 to 48 points, with higher scores indicating better clinical reasoning performance. |
| Study-Specific Clinical Case Analysis Assessment: AI Critical Appraisal Component Score | Immediately after completion of the educational intervention | AI critical appraisal ability was assessed using Case 3 of the study-specific Clinical Case Analysis Assessment. Students evaluated an AI-generated treatment plan for a patient with acute kidney injury and severe hyperkalemia, identified clinically inappropriate or unsafe recommendations, explained the reasons, prioritized immediate management, and proposed an appropriate treatment plan. Scores ranged from 0 to 22 points, with higher scores indicating better ability to critically appraise AI-generated clinical recommendations. |
| Knowledge Gain in Theoretical Knowledge Test Score | Change from immediately before the educational intervention to immediately after completion of the educational intervention | Knowledge gain was calculated as the post-class theoretical knowledge test score minus the pre-class theoretical knowledge test score. Both tests consisted of 15 single-choice questions and had possible total scores ranging from 0 to 30 points. Therefore, knowledge gain scores could range from -30 to +30 points. Higher values indicate greater improvement in theoretical knowledge, a score of 0 indicates no change, and negative values indicate a lower post-class than pre-class score. |
| Study-Specific Course Feedback Questionnaire Total Score | Immediately after completion of the educational intervention | Course feedback was assessed using a study-specific 8-item questionnaire developed by the research team. Items assessed perceived learning gains and teaching satisfaction. Each item was rated on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Total scores ranged from 8 to 40 points, with higher scores indicating more favorable perceptions of the educational experience. |
Countries
China
Contacts
PRINCIPAL_INVESTIGATORFulin Dou
The Second Hospital of Shandong University
Outcome results
None listed