Novel nomogram and decision curve analysis for predicting head and neck skin cancer risk

In Kashgar, Xinjiang, China, skin cancer accounts for 51% of head and neck malignant tumors, and the incidence rate of head and neck skin cancer in this region is much higher than the national average level. This current situation highlights the necessity of risk prediction. This study retrospectively analyzed 1,156 participants from the 2015-2024 Kashgar Facial Skin Health Survey. The study employed logistic regression to screen for risk factors, and through sensitivity analysis, finally identified 8 factors that are independently associated with a high risk of skin cancer. This study developed a nomogram model for head and neck skin cancer incorporating a variety of risk factors. Key findings revealed that individuals with frequent cosmetics use, long-term high-fat diet, inadequate vegetable intake, lip-biting habit, scratching habit, smoking behavior, prolonged outdoor exposure, and those who do not wear hats have a significantly increased risk of head and neck skin cancer. The model was evaluated using the Receiver Operating Characteristic (ROC) curve (with Area Under the Curve, AUC), calibration curves, and Decision Curve Analysis (DCA). The results showed that the model exhibited good reliability and accuracy in both groups, and possessed high clinical value within specific threshold ranges. In conclusion, this nomogram model can assist clinicians in identifying high-risk individuals for head and neck skin cancer, and provide guidance for the prevention efforts of head and neck skin cancer.