Development and Application of an AI Chatbot for Perioperative Health Education in Thyroid Cancer Patients

Condition(s) studied

Investigational drug(s) / intervention(s)

Routine Perioperative Health Education: Participants in this group will receive standard, traditional perioperative health education. The intervention is delivered orally by three certified senior oncology nurses at 5 predefined clinical timeline milestones. A standardized health education protocol is followed, which comprehensively covers general information about thyroid cancer, pre-admission preparation, surgical day and postoperative care, symptom management, discharge instructions, medication adherence guidance, and transitional care. After hospital discharge, participants will receive routine follow-up support through the hospital's official mini-program platform's health consultation portal.

Coze-Based AI Chatbot Health Education: In addition to the routine health education received by the control group, participants in this group will interact with a novel, interactive health education chatbot developed on the "Coze" AI platform. The chatbot intervention framework consists of three core components: User Guide \& Free Inquiry: Patients receive instructions on how to use the AI agent and can initiate unstructured, free-text inquiries in natural language based on their individual needs. Proactive Health Education Push: The chatbot automatically pushes targeted health education modules to patients at predefined clinical milestones throughout the perioperative journey. Interactive Precaution Explanations: The chatbot provides detailed, real-time explanations of postoperative precautions and self-care guidelines. Furthermore, a dedicated technical and clinical maintenance team will continuously optimize the system's prompt words and interaction designs based on patient feedback and user behavioral analysis to enha

Study summary

The goal of this clinical trial is to evaluate the usefulness and effectiveness of a newly developed artificial intelligence (AI) chatbot for health education in thyroid cancer patients during the period around their surgery (the perioperative period). The main questions it aims to answer are:

Does the AI chatbot improve patients' readiness for discharge, quality of recovery, and active engagement in their own healthcare compared to traditional education methods?

How do patients rate the usability and friendliness of this new AI chatbot system?

Researchers will compare a group of patients who receive traditional health education (the control group) with a subsequent group of patients who use the new AI chatbot built on the "Coze" platform (the intervention group) to see if the AI chatbot leads to better recovery experiences and higher patient engagement.

Participants will be asked to:

Access and interact with the AI chatbot to receive personalized health guidance, ask questions, and get instant feedback during their hospital stay.

Complete a set of questionnaires on the morning of their hospital discharge to evaluate their recovery quality, readiness to leave the hospital, and level of engagement in their care.

Complete an extra survey regarding the usability and acceptability of the chatbot system (only for participants in the AI chatbot group).

Eligibility

Primary outcome measure(s)

• Readiness for Hospital Discharge — On the morning of hospital discharge
  Measured using the Readiness for Hospital Discharge Scale (RHDS). The scale consists of 23 items across four dimensions: personal status (7 items), knowledge of the disease (8 items), coping ability (3 items), and expected support (4 items). Each item is scored on a 0-10 scale. The total score ranges from 0 to 220, with higher scores indicating a higher and more sufficient readiness for hospital discharge. The overall Cronbach's alpha of the scale is 0.97.
• Postoperative Quality of Recovery — On the morning of hospital discharge
  Evaluated using the Chinese version of the 15-item Quality of Recovery Scale (QoR-15). It measures the patient's early postoperative recovery quality across five dimensions: physical comfort, independence, psychological support, emotional state, and pain. The total score ranges from 0 to 150, with higher scores representing a better and higher quality of recovery.
• System Usability of the AI Chatbot — On the morning of hospital discharge.
  Measured exclusively in the intervention group using the System Usability Scale (SUS) to evaluate the overall usability and user satisfaction of the chatbot. The scale consists of 10 items covering three dimensions: effectiveness, learnability, and satisfaction. The total score ranges from 0 to 100, with higher scores indicating better system usability. A total score greater than 60 is generally considered acceptable and indicates good usability. The content validity index (CVI) of the scale is 0.822.

Trial sites (1)

More Tianjin Medical University Cancer Institute and Hospital trials in China

Other trials for the same condition