Development and validation of a prediction model for psychological distress in patients with differentiated thyroid cancer undergoing ¹³¹I therapy.

OBJECTIVE

To develop and validate a risk prediction model for psychological distress in patients with differentiated thyroid cancer (DTC) undergoing ¹³¹I therapy.

METHODS

A total of 247 DTC patients who received ¹³¹I therapy at a tertiary Grade A hospital in Taiyuan between May and December 2025 were enrolled. Independent risk factors for psychological distress were identified to construct the model. Model performance was evaluated using receiver operating characteristic (ROC) curves, the Hosmer-Lemeshow test, and calibration curves.

RESULTS

Multivariate logistic regression identified poor sleep quality, thyroid-stimulating hormone (TSH) stimulation, pulmonary metastasis, social support, radiation exposure concerns, and lower serum 25-hydroxyvitamin D levels as independent risk factors (all P < 0.05). The nomogram model achieved an area under the ROC curve (AUC) of 0.835, with a sensitivity of 83.1% and a specificity of 68.2%. The Hosmer-Lemeshow test yielded a χ² of 12.064 (P = 0.148). Calibration curves and decision curve analysis (DCA) confirmed good calibration and clinical utility. Internal validation via bootstrap resampling and 10-fold cross-validation produced AUCs of 0.82 and 0.77, respectively.

CONCLUSION

This internally validated model is a promising tool for predicting psychological distress in DTC patients undergoing ¹³¹I therapy. However, external validation in larger, multicenter cohorts is necessary before routine clinical use can be recommended.