AI-Assisted Education Before Surgery to Prevent Lung Complications in Older Surgical Patients

This single-center, randomized controlled trial will evaluate the effect of artificial intelligence-assisted preoperative patient education on postoperative pulmonary complications in patients aged 65 years and older undergoing major abdominal surgery at Erciyes University Health Application and Research Center in Kayseri, Türkiye.

Educational content will be developed using ChatGPT and converted into an illustrated patient education material. Before clinical use, the material will be assessed for readability and reviewed by 10 surgical nursing experts using the Patient Education Materials Assessment Tool for Printable Materials (PEMAT-P).

Participants will be assigned to the intervention or control group using a computer-generated simple randomization sequence. A researcher will obtain written informed consent and deliver the educational intervention. A separate researcher, unaware of group allocation, will collect baseline data and assess postoperative complications.

Participants in the intervention group will receive preoperative education at least 24 hours before surgery using the prepared educational material. Their questions will be answered, and a printed copy of the material will be provided. Participants in the control group will receive standard nursing care without the study-specific educational intervention.

Baseline demographic and clinical characteristics will be recorded, including preoperative oxygen saturation and the ARISCAT score for postoperative pulmonary risk assessment. Postoperative pulmonary and other surgical complications will be recorded using a structured complication follow-up form on postoperative days 1, 3, 7, and 30. Assessments will be conducted face-to-face during hospitalization and by telephone after discharge.

• Artificial Intelligence-Assisted Preoperative Patient Education (Behavioral): A researcher will provide face-to-face patient education at least 24 hours before major abdominal surgery using an illustrated educational material developed with the assistance of ChatGPT and reviewed by surgical nursing experts before use. The education will focus on preventing postoperative pulmonary complications. Participants' questions will be answered, and a printed copy of the material will be provided for further reference.

Primary outcomes

• Proportion of Participants With at Least One Postoperative Pulmonary Complication [Within 30 days after surgery]: The proportion of participants who develop at least one postoperative pulmonary complication within 30 days after surgery. Events recorded on the researcher-developed Postoperative Complication Follow-up Form include atelectasis, pneumonia, respiratory failure, pleural effusion, pneumothorax, pulmonary embolism, pulmonary edema, reintubation, acute respiratory distress syndrome, aspiration pneumonia, respiratory tract infection, and bronchospasm. The postoperative day of occurrence will be recorded. Assessments will be performed by a researcher blinded to group allocation on postoperative days 1, 3, 7, and 30, through face-to-face interviews during hospitalization and telephone interviews after discharge.Participants' reported symptoms will be evaluated by a physician, who will determine whether a postoperative pulmonary complication is present.

Sex: All, ages: 65 Years to any

Inclusion Criteria:

* Aged 65 years or older.
* Able to read and write.
* Scheduled for elective major abdominal surgery for colorectal, gastric, or pancreatic cancer.
* Willing to participate and provide written informed consent.

Exclusion Criteria:

* Undergoing reoperation due to a surgical complication.
* Communication difficulties associated with dementia, Alzheimer's disease, or intellectual disability.
* Visual impairment.
* Hearing impairment.
* Unable to speak Turkish.

• Hilal KAHRAMAN, Principal investigator, TC Erciyes University