To develop and validate a predictive model for identifying risk factors associated with postoperative sleep disturbances in patients with oral cancer, emphasizing its clinical significance.
Key Findings:
Postoperative sleep disturbances occurred in 39.41% of the training set and 39.66% of the validation set, indicating a significant prevalence.
The model demonstrated excellent discrimination with AUCs of 0.902 (training) and 0.967 (validation), suggesting high predictive accuracy.
Interpretation:
The developed nomogram effectively identifies oral cancer patients at risk for postoperative sleep disturbances, highlighting the importance of early intervention strategies.
Limitations:
The study's reliance on convenience sampling may limit generalizability and introduce potential biases.
The single-center design may not reflect broader population characteristics.
Conclusion:
A practical predictive model was established, demonstrating strong discrimination and clinical utility for identifying patients at risk of postoperative sleep disturbances, with implications for future research and clinical practice.
