Abstract
OBJECTIVE: The substantial prevalence of nonadherence to analgesic medication among individuals diagnosed with cancer imposes a significant strain on both patients and healthcare resources. The objective of this study is to develop and authenticate a nomogram model for assessing nonadherence to analgesic medication in cancer patients. METHODS: Clinical information, demographic data, and medication adherence records of cancer pain patients were gathered from the Affiliated Hospital of Chengde Medical University between April 2020 and March 2023. The risk factors associated with analgesic medication nonadherence in cancer patients were analyzed using the least absolute selection operator (LASSO) regression model and multivariate logistic regression. Additionally, a nomogram model was developed. The bootstrap method was employed to internally verify the model. Discrimination and accuracy of the nomogram model were evaluated using the Concordance index (C-index), area under the receiver Operating characteristic (ROC) curve (AUC), and calibration curve. The potential clinical value of the nomogram model was established through decision curve analysis (DCA) and clinical impact curve. RESULTS: The study included a total of 450 patients, with a nonadherence rate of 43.33%. The model incorporated seven factors: age, address, smoking history, number of comorbidities, use of nonsteroidal antiinflammatory drugs (NSAIDs), use of opioids, and PHQ-8. The C-index of the model was found to be 0.93 (95% CI: 0.907-0.953), and the ROC curve demonstrated an AUC of 0.929. Furthermore, the DCA and clinical impact curves indicate that the built model can accurately predict cancer pain patients' medication adherence performance. CONCLUSIONS: A nomogram model based on 7 risk factors has been successfully developed and validated for long-term analgesic management of cancer patients.