A nomogram for predicting the risk of cancer-related cognitive impairment in breast cancer patients based on a scientific symptom model
Zhongtao Zhou,Jiajia Ren,Qiankun Liu,Shuoshuo Li,Jiahui Xu,Xiaoyan Wu,Yuanxiang Xiao,Zipu Zhang,Wanchen Jia,Huaiyu Bai,Jing Zhang
DOI: https://doi.org/10.1038/s41598-024-65406-5
IF: 4.6
2024-06-26
Scientific Reports
Abstract:Cancer-related cognitive impairment is a significant clinical challenge observed in patients with breast cancer, manifesting during or after treatment. This impairment leads to deteriorations in memory, processing speed, attention, and executive functioning, which profoundly impact patients' occupational performance, daily living activities, and overall quality of life. Grounded in the Symptom Science Model 2.0, this study investigates the contributing factors to Cancer-related cognitive impairment in breast cancer patients and develops a predictive nomogram for this demographic. Employing both univariate and multivariate logistic regression analyses, this investigation delineates the predictive factors influencing outcomes in breast cancer patients. A nomogram was constructed leveraging these identified predictive factors, accompanied by internal validation through bootstrap resampling methodology (1000 bootstrap samples). The efficacy of the predictive model was assessed by employing the Hosmer–Lemeshow goodness-of-fit test and calibration curves. The prevalence of cognitive impairment in breast cancer patients was identified to be 45.83%.Multivariate logistic regression analysis identified the independent predictors of Cancer-related cognitive impairment in breast cancer patients as place of residence, educational level, chemotherapy, benefit finding, post-traumatic growth, anxiety, fear of cancer progression, and fasting blood glucose levels. these factors were integrated into the nomogram. The Hosmer–Lemeshow goodness-of-fit test demonstrated that the prediction model was appropriately calibrated (χ 2 = 11.520, P = 0.174). Furthermore, the model exhibited an area under the curve of 0.955 (95% CI 0.939 to 0.971) and a sensitivity of 0.906, evidencing its robust discriminative capacity and accuracy. Utilizing the Symptom Science Model 2.0 as a framework, this study comprehensively examines the multifaceted factors influencing Cancer-related cognitive impairment in breast cancer patients, spanning five critical domains: complex symptoms, phenotypic characterization, biobehavioral factors, social determinants of health, and patient-centered experiences. A predictive nomogram model was established, demonstrating satisfactory predictive accuracy and capability. This model is capable of identifying breast cancer patients with cognitive impairments with high precision. The findings furnish empirical evidence in support of the early detection, diagnosis, and intervention strategies for high-risk breast cancer patients afflicted with Cancer-related cognitive impairment.
multidisciplinary sciences