Objectives: Depression is common in hepatocellular carcinoma patients. Predicting depression is critical for managing hepatocellular carcinoma patients. The aim of this study was to establish a comprehensive visual predictive nomogram for calculating a probabilistic estimate of depression in hepatocellular carcinoma patients.Methods: Two-hundred and sixty-nine patients diagnosed with hepatocellular carcinoma were included. The demographic and clinical information was collected from medical records. A nomogram was constructed based on the results of a multivariate regression analysis. Discrimination and calibration were used to assess its performance. A decision curve analysis was used to assess its clinical utility.Results: One hundred and thirty-four (49.82%) hepatocellular carcinoma patients with depression were identified. The education level, social support, pain intensity, disclosure of the cancer diagnosis to the patient and Child-Pugh class were included in the nomogram. The AUC was 0.828, indicating good discrimination. The nomogram demonstrated calibration curves with slight deviation from the ideal predictions and a high net benefit in the decision curve analysis.Conclusions: We proposed a nomogram for the individualised prediction of depression in hepatocellular carcinoma patients with satisfactory performance and clinical utility.Key points/highlightsA visual, comprehensive and convenient nomogram was established for the prediction of depression in patients with hepatocellular carcinoma.The proposed nomogram showed satisfactory validity, discrimination and clinical utility, indicating good performance for predicting depression.The variables found to be sufficiently informative and predictive to warrant inclusion in the nomogram were the patient education level, pain intensity, social support, disclosure of the hepatocellular carcinoma diagnosis to the patient and Child-Pugh class.
Keywords: Hepatocellular carcinoma; cancer; cross-sectional study; depression; nomogram.