邱曉彥徐雯娟2026-09-092026-09-092026-07-13https://handle.ncl.edu.tw/11296/tm4n8whttps://203.71.86.71/handle/123456789/74490學位別：博士 語文別：英文 指導教授：邱曉彥 口試委員：蔡佩珊; 黃惠娟; 高煌凱; 謝萬雲; 邱曉彥 授權範圍：開放校內, 開放日期為2026-07-28;校外, 開放日期為2028-07-27 電子論文連結：https://handle.ncl.edu.tw/11296/tm4n8wBackground: Dysphagia is one of the most common and debilitating complications following oral cancer surgery with free-flap reconstruction. It adversely affects oral intake, nutritional status, psychological well-being, and quality of life, while increasing the risk of postoperative complications and delayed recovery. Despite growing evidence on postoperative dysphagia, important gaps remain in the early identification of high-risk patients, disease-specific assessment, comparative evaluation of swallowing rehabilitation strategies, and the application of digital health technologies to support long-term rehabilitation. Consequently, an integrated evidence base to guide dysphagia assessment and management in this population is lacking. Purposes: This dissertation aimed to establish an evidence base for developing a nursing care model for the assessment and management of dysphagia in patients with oral cancer undergoing free-flap reconstruction. Four complementary studies were conducted to (1) investigate the occurrence, risk factors, and clinical impact of early postoperative dysphagia; (2) validate a disease-specific assessment instrument for postoperative oral dysfunction; (3) compare the effectiveness of swallowing rehabilitation strategies through network meta-analysis; and (4) evaluate an AI-based swallowing rehabilitation program. Methods: This dissertation comprised four complementary studies using retrospective cohort, psychometric validation, systematic review with network meta-analysis, and randomized controlled trial designs. Study 1 examined the epidemiology and clinical outcomes of postoperative dysphagia in 892 patients. Study 2 evaluated the psychometric properties of the Traditional Chinese version of the Postoperative Oral Dysfunction Scale-10 (TC-POD-10) in 100 patients. Study 3 synthesized evidence from randomized controlled trials comparing swallowing rehabilitation interventions for patients with head and neck cancer. Study 4 evaluated a 12-week AI-based swallowing rehabilitation program in a randomized controlled trial involving 50 patients with oral cancer undergoing free-flap reconstruction. Results: Study 1 demonstrated that postoperative dysphagia occurred in 76.1% of patients. Lower body mass index, smoking, alcohol consumption, betel nut chewing, higher postoperative pain, neck lymph node dissection, and preoperative adjuvant therapy were identified as independent risk factors. Patients with dysphagia experienced longer hospital stays and higher rates of nasogastric tube placement, tracheostomy, and 30-day readmission. Study 2 revealed that the TC-POD-10 had excellent psychometric properties, including a one-factor structure explaining 65.6% of the total variance, excellent internal consistency (Cronbach's α = 0.94), and high diagnostic accuracy (AUC = 0.92, sensitivity 89%, specificity 96%), supporting an optimal screening cutoff score of 15 for postoperative dysphagia. Study 3 synthesized evidence from 14 randomized controlled trials involving 1,128 participants and showed that combined swallowing exercises with nutritional support produced the greatest improvement in swallowing function compared with routine care (SMD = −0.71, 95% CI = −1.09 to −0.33), whereas nutrition intervention demonstrated the greatest improvement in swallowing-related quality of life (1.94, 0.77–3.11). Technology-assisted swallowing rehabilitation also significantly improved swallowing function compared with routine care (−0.32, −0.63 to −0.01). Study 4 demonstrated that the AI-based swallowing rehabilitation program significantly improved postoperative oral dysfunction ( p < 0.001 at week 4 and 12), oral intake function (p <0.05 at week 4 and 12), anxiety (p <0.05 at week 4 and 12), and quality of life (p < 0.001 at Week 12) compared with usual care. Rehabilitation adherence was also significantly higher in the intervention group at Weeks 4 and 12 (p = 0.030 and 0.006, respectively). Conclusions: This dissertation establishes a comprehensive evidence base for dysphagia assessment and management following oral cancer surgery with free-flap reconstruction. By integrating epidemiological evidence, disease-specific assessment, comparative effectiveness evidence, and digital health intervention, this research provides a systematic framework for evidence-based dysphagia care. The proposed nursing care model supports early risk identification, standardized assessment, evidence-based rehabilitation, and long-term self-management, with the potential to improve postoperative swallowing recovery and quality of life in patients with oral cancer. Collectively, these four studies establish a translational pathway from risk identification to assessment, evidence synthesis, and technology-assisted intervention, providing a foundation for developing and implementing evidence-based nursing care for dysphagia following oral cancer surgery.系所名稱：護理學系博士班thesis