Influencing factors and predictive model construction of anxiety and depression in patients with cervical cancer

Anxiety and depression are highly prevalent among patients with cervical cancer (CC). However, few studies have systematically analyzed the psychological effects of tumor stage, treatment methods, and related factors on these patients, or developed predictive models for these outcomes.

To identify factors influencing anxiety and depression in patients with CC and construct predictive models.

We retrospectively analyzed data from 119 patients with CC treated at the Gynecology Department of Suzhou Ninth People's Hospital between January 2017 and May 2025. Clinical data, psychological hope levels at diagnosis, and Self-Rating Anxiety Scale and Self-Rating Depression Scale scores during treatment were collected. Influencing factors were identified, and predictive models were developed. The model performance was evaluated using receiver operating characteristic (ROC) curves and the Hosmer-Lemeshow goodness-of-fit test.

During treatment, 64.71% of the patients experienced anxiety and 52.10% experienced depression. Significant differences in family income, tumor stage, treatment modality, and hope level were observed between patients with and without anxiety/depression (P < 0.05). Multivariate analysis showed that a family monthly income < 5000 yuan, stage III-IV tumor, comprehensive treatment, and low hope level were independent risk factors (P < 0.05). The predictive formula for anxiety was as follows: Logit (P) = 0.795 × monthly income + 0.594 × tumor stage + 1.095 × treatment method + 1.184 × hope level - 9.176; for depression: Logit (P) = 0.432 × monthly income + 0.518 × tumor stage + 0.727 × treatment method + 1.095 × hope level - 8.541. The area under the ROC curves were 0.865 for anxiety and 0.837 for depression. Goodness-of-fit test confirmed no overfitting (P > 0.05).

Family income, tumor stage, treatment method, and hope level are key determinants of anxiety and depression in patients with CC. Predictive models incorporating these factors can effectively assess risk of anxiety and depression during treatment.