Enhancing hospice care with psychological support and a nomogram to predict delirium in patients with advanced solid tumors

Abstract

To assesses the impact of integrating hospice care with psychological interventions on patient well-being and to introduce a predictive nomogram model for delirium that incorporates clinical and psychosocial variables, thereby improving the accuracy in hospice care environments. Data from 381 patients treated from September 2018 to February 2023 were analyzed. The patients were divided into a control group (n=177, receiving standard care) and an experimental group (n=204, receiving combined hospice care and psychological interventions) according to the treatment modality. The duration of care extended until the patient's discharge from the hospital or death. The experimental group demonstrated significant improvements in emotional well-being and a lower incidence of delirium compared to the control group. Specifically, emotional well-being assessments revealed marked improvements in the experimental group, as evidenced by lower scores on the Self-Rating Anxiety Scale (SAS) and Self-Rating Depression Scale (SDS) post-intervention. The nomogram model, developed using logistic regression based on clinical characteristics, effectively predicted the risk of delirium in patients with advanced cancer. Significant predictors in the model included ECOG score ≥3, Palliative Prognostic Index score ≥6, opioid usage, polypharmacy, infections, sleep disorders, organ failure, brain metastases, electrolyte imbalances, activity limitations, pre-care SAS score ≥60, pre-care SDS score ≥63, and pre-care KPS score ≥60. The model's predictive accuracy was validated, showing AUC values of 0.839 for the training cohort and 0.864 for the validation cohort, with calibration and Decision Curve Analysis (DCA) confirming its clinical utility. Integrating hospice care with psychological interventions not only significantly enhanced the emotional well-being of advanced cancer patients but also reduced the actual incidence of delirium. This approach, offering a valuable Nomogram model for precise care planning and risk management, underscores the importance of integrated, personalized care strategies in advanced cancer management.