CABG-AI-Supported Discharge Education

The Effect of Artificial Intelligence-Supported Individualized Discharge Education on Postoperative Pain, Anxiety, Stress Response, Recovery, and Discharge Education Satisfaction in Patients Undergoing Coronary Artery Bypass Graft Surgery: A Randomized Controlled Trial

This randomized controlled trial will evaluate the effectiveness of artificial intelligence-supported individualized discharge education for patients undergoing coronary artery bypass graft (CABG) surgery. Participants will be assigned to either the intervention group, which will receive personalized AI-generated discharge education delivered through a nurse-avatar video in addition to standard verbal education, or the control group, which will receive standard discharge education only. The study will compare postoperative pain, anxiety, physiological stress response, recovery, and satisfaction with discharge education between the two groups. Outcomes will be measured during hospitalization and at the routine postoperative follow-up visit. The findings may help improve patient education methods and postoperative recovery in cardiac surgery patients.

Coronary artery bypass graft (CABG) surgery is a major cardiac procedure that requires structured discharge education to support patients' physical, psychological, and behavioral recovery after hospitalization. Traditional discharge teaching provided by nurses is often limited by time constraints, varying patient needs, and inconsistent information delivery. Recent advances in artificial intelligence (AI) offer new opportunities to provide individualized, understandable, and repetitive patient education that may enhance postoperative recovery.

This randomized controlled trial aims to evaluate the effectiveness of AI-supported individualized discharge education for patients undergoing CABG surgery. The intervention consists of a personalized discharge education script generated by an advanced large-language-model system (ChatGPT), structured according to 6 main topics and 24 sub-domains, and then converted into a nurse-avatar video. Patients in the intervention group will receive this AI-supported education in addition to the standard verbal discharge education provided by cardiovascular surgery nurses. Patients in the control group will receive only the standard discharge education based on current hospital protocols.

The study will assess multiple postoperative outcomes, including pain intensity, anxiety levels, physiological stress response (cortisol, adrenaline, noradrenaline, ACTH, glucose, and insulin), recovery after discharge using a validated surgical recovery scale, and satisfaction with discharge education. Measurements will be collected at four time points: the day before discharge, the morning of discharge before education, immediately after the completion of discharge education, and at the routine postoperative follow-up visit.

A total of 128 participants will be randomized into two groups using block randomization stratified by sex. All data will be collected using validated scales and standardized laboratory procedures. This study is expected to provide evidence on whether AI-supported discharge education can improve postoperative recovery, reduce stress and anxiety, and increase satisfaction among CABG patients. The results may contribute to the development of innovative, technology-enhanced education models for surgical patients and support the integration of AI tools into nursing practice.

Inclusion Criteria:

• Patients who have undergone coronary artery bypass graft (CABG) surgery.
• Age 18 years or older.
• Hemodynamically stable at the time of enrollment.
• Able to understand and communicate in Turkish.
• Able to use or have access to a mobile phone capable of viewing the educational video.
• Able to provide informed consent.
• Scheduled for routine postoperative follow-up at the study hospital.

Exclusion Criteria:

• Patients undergoing emergency CABG surgery.
• Presence of severe cognitive impairment that prevents understanding or following instructions.
• Diagnosed psychiatric disorders (e.g., schizophrenia, severe depression) that interfere with study participation.
• Severe vision or hearing impairment that prevents receiving the education.
• Presence of postoperative complications requiring intensive care readmission.
• Patients with endocrine disorders affecting stress biomarkers (e.g., Cushing's syndrome, Addison's disease).
• Chronic use of corticosteroids or medications that significantly alter cortisol levels.
• Patients who decline to participate or withdraw consent.
• Patients who are unable to complete follow-up assessments.