Effectiveness of Artificial Intelligence Based Educational Approach in Developing Clinical Reasoning Skills

Not Applicable

Recruiting

• Conditions: Student; Student Engagement; Student Education

• Registration Number: NCT07010991
• Lead Sponsor: Medipol University

Brief Summary

In the context of this study, the 2nd and 3rd grade undergraduate students in Medipol University, Department of Physiotherapy and Rehabilitation, will taken into an education program to increase and assess their theoretical knowledge level, clinical decision-making skills and attitudes towards interactive learning with artificial intelligence by applying the ChatGPT-supported PBL module and the results will be compared with traditional teaching methods.

Detailed Description

In recent years, the use of artificial intelligence-based applications in the field of health education has been rapidly increasing; especially large language models such as ChatGPT have attracted attention in terms of providing rapid access to information and individualized learning opportunities. Utilizing this potential of artificial intelligence, especially in active learning methods such as problem-based learning (PBL), is seen as an approach that can strengthen student-centered education. It is reported in the literature that ChatGPT-supported learning applications yield positive results in medical education, increase theoretical success and provide significant improvements in clinical skills. However, applications specific to physiotherapy education are still limited in this regard. Physiotherapy education requires students to have not only theoretical knowledge but also versatile skills such as clinical decision-making, problem-solving, and patient communication. Therefore, it is important to systematically investigate the effect of ChatGPT-supported PBL applications on student performance and satisfaction. Professionals should be able to benefit from these systems when they do not know where to start in a different case, and students should be informed about this issue by academicians who are experts in their fields in the university environment. In order to prevent the misuse of these systems, it is thought that academicians should not avoid technology and should benefit from the interactive course processing and correct communication opportunities that these systems can create with patients. Today, teaching students how to use technology instead of telling them not to use it is very important for academic success. In this context, the aim of this study is to evaluate the student's theoretical knowledge level, clinical decision-making skills and attitudes towards interactive learning with artificial intelligence and compare them with traditional teaching methods by applying the ChatGPT-supported PBL module to volunteer groups consisting of undergraduate students of Istanbul Medipol University, Department of Physiotherapy and Rehabilitation.

Study Timeline

• Status Verified: 2025-05
• Study Start: 2025-05-14
• Study First Submitted: 2025-05-30
• Study First Submitted QC: 2025-05-30
• Last Update Submitted: 2025-05-30
• Study First Posted: 2025-06-08
• Last Update Posted: 2025-06-08
• Primary Completion: 2025-08-31
• Study Completion: 2025-09-10

Recruitment & Eligibility

• Status: RECRUITING
• Sex: All
• Target Recruitment: 40

Inclusion Criteria

• Being an undergraduate student
• Being voluntary to attend to the courses

Exclusion Criteria

• Not attending to the sessions on time
• Being a 4th grade student who already took and learned the process of clinical problem solving
• Want to recruit to study during the study process

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: PARALLEL

Primary Outcome Measures

Name	Time	Method
Theoretical knowledge test	4 weeks	A 20-question multiple choice exam will be administered at the beginning of the study after the initial theoretical LBP course and at the end of the 2nd week of training, in total 3 times. With this way, distinguishing the effect of the lecturer's initial education and student's own studies and research will be possible at the analysis. The exams will be compared between groups to assess the intervention effect. The maximum score for the exam will be 100, and a score below 80 will be considered a failing grade.
Mini Clinical Evaluation Exam (Mini-CEX)	4 weeks	Participant's history taking, physical assessment and clinical reasoning skills will be observed by an independent assessor using the 9-point Mini-CEX form. Mini-CEX has been widely accepted as an effective and reliable method for assessing clinical skills. It includes student's ability to obtain a patient's history from the patient and family and to perform physical examinations under the guidance of an instructor. The Mini-CEX scoring system uses a nine-point scale and includes seven standards: medical interview skills, physical examination skills, humanistic qualities/professionalism, clinical judgment, counseling skills, organizational efficiency and general clinical competence. The scale ranges from inadequate (1-3 points), satisfactory (4-6 points) to superior (7-9 points). To maintain consistency, all Mini-CEX assessments will be made by a single assessor. The scale is valid and reliable in Turkish.

Secondary Outcome Measures

Name	Time	Method
Student Satisfaction and Attitude Survey	4 weeks	Each student will be asked to complete a self-assessment form, where they evaluate their contributions to the group dynamics and their individual learning process, and a peer-assessment form, where they rate the participation levels of their group mates. Students' satisfaction and the impact of using PBL-ChatGPT on their learning experiences will be assessed with a survey developed from a previous study.
Young Internet Addiction Test	4 weeks	This scale, which is used to determine the internet usage habits of the participants, was developed by Young in 1998. There is also a Turkish version of the short form consisting of 12 items for quick assessment. The psychometric properties of the questionnaire were studied by Kutlu et al. and it is valid and reliable in Turkish. Participants are scored between 12-60, with score 1 indicating never and score 5 indicating always. High scores indicate a high risk of internet addiction.
Adult Reading Motivation Scale	4 weeks	The scale, developed by Schutte and Malouff (2007) to determine adults' reading motivation, consists of 19 items. The scale assesses individuals' reading motivation by asking questions on subheadings such as self, recognition of competence, and reading to be successful in other areas. A score of 1 indicates strongly disagree, a score of 5 indicates strongly agree, and participants receive scores between 19 and 95. High scores indicate high reading motivation and habit. The psychometric properties of the survey were conducted by Yıldız et al. and it is valid and reliable in Turkish.
Artificial Intelligence Self-Efficacy Scale	4 weeks	The Artificial Intelligence Self-Efficacy Scale (AISES) will be used to measure participants' perceptions of competence in using artificial intelligence technologies. The scale was developed by Wang and Chuang (2023) and adapted to Turkish by Uyan and Gültekin (2024) to test its validity and reliability. In the Turkish adaptation study, it was confirmed that the scale consisted of four sub-dimensions (assistance, anthropomorphic interaction, comfort and technological competencies) and contained a total of 21 items. The goodness of fit values were found to be at an acceptable level (CFI=0.887, RMSEA=0.084) and Cronbach's alpha values ranged between 0.708 and 0.782 for the sub-dimensions. The scale to be applied to the participants was structured as a 5-point Likert type (1 = Strongly Disagree, 5 = Strongly Agree).

Trial Locations

Locations (1)

Medipol University İstanbul; 🇹🇷 İstanbul, Kavacık Beykoz, Turkey