Scholarlink

Objective <p>The aim of this study was to construct and internally validate a prediction model for the risk of preoperative frailty in Chinese breast cancer patients.</p> Method <p>From October 2022 to August 2023, 519 patients were selected who presented for elective surgery in a tertiary hospital of breast surgery in Jinzhou City, China. Patients were preoperatively identified as frail or not frail, using standardized criteria. Binary logistic regression analysis was used to screen the risk factors for the occurrence of preoperative frailty in these breast cancer patients, to establish a risk prediction model. The discriminatory nature of the constructed nomogram model was evaluated with the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve. After internal validation of the nomogram model, correction curves were produced using the bootstrap method with 1000 repetitive sampling of the raw data, and the accuracy of the model was evaluated by the Hosmer-Lemesow goodness-of-fit test. Finally, clinical decision curve analysis (DCA) was used to evaluate the model.</p> Results <p>Logistic regression analysis identified 7 variables associated with frailty in the study cohort that were entered into the model: chemotherapy, serum levels of albumin (ALB) and hemoglobin (HB), nutrition status, occurrence of depression, social support, and Instrumental Activities of Daily Living (IADL) scores. Of these, protective factors were nutrition and social support, while the identified risk factors were chemotherapy, ALB, HB, depression, and IADL. The model had an AUC value of 0.87 (95% CI: 0.837 ~ 0.904). The outcome of the Hosmer-Lemeshow test was χ<sup>2</sup> = 0.521, <i>P</i> = 0.771, indicating predictions were not statistically different from the actual data. The calibration curve also showed consistency between predicted and actual probability. The ROC and DCA indicated that the nomogram model had a good theoretical predictive performance.</p> Conclusion <p>The predictive model of breast cancer preoperative frailty risk obtained in this study is expected to have a good predictive ability, with satisfying predicted levels of differentiation, accuracy and practicality. The model provides a basis for clinical nursing staff to screen and timely detect breast Chinese cancer patients at risk of preoperative frailty.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Constructing and evaluating a predictive model for the risk of preoperative frailty in breast cancer patients

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月28日

查看全文
  • Qianna Fang,
  • Hongmei Jiang,
  • Lingfang Deng,
  • Weiming Sun,
  • Mei Wang,
  • Yu Liu,
  • Jinjiang Xu

摘要

Objective

The aim of this study was to construct and internally validate a prediction model for the risk of preoperative frailty in Chinese breast cancer patients.

Method

From October 2022 to August 2023, 519 patients were selected who presented for elective surgery in a tertiary hospital of breast surgery in Jinzhou City, China. Patients were preoperatively identified as frail or not frail, using standardized criteria. Binary logistic regression analysis was used to screen the risk factors for the occurrence of preoperative frailty in these breast cancer patients, to establish a risk prediction model. The discriminatory nature of the constructed nomogram model was evaluated with the Area Under the Curve (AUC) of the Receiver Operating Characteristic (ROC) curve. After internal validation of the nomogram model, correction curves were produced using the bootstrap method with 1000 repetitive sampling of the raw data, and the accuracy of the model was evaluated by the Hosmer-Lemesow goodness-of-fit test. Finally, clinical decision curve analysis (DCA) was used to evaluate the model.

Results

Logistic regression analysis identified 7 variables associated with frailty in the study cohort that were entered into the model: chemotherapy, serum levels of albumin (ALB) and hemoglobin (HB), nutrition status, occurrence of depression, social support, and Instrumental Activities of Daily Living (IADL) scores. Of these, protective factors were nutrition and social support, while the identified risk factors were chemotherapy, ALB, HB, depression, and IADL. The model had an AUC value of 0.87 (95% CI: 0.837 ~ 0.904). The outcome of the Hosmer-Lemeshow test was χ2 = 0.521, P = 0.771, indicating predictions were not statistically different from the actual data. The calibration curve also showed consistency between predicted and actual probability. The ROC and DCA indicated that the nomogram model had a good theoretical predictive performance.

Conclusion

The predictive model of breast cancer preoperative frailty risk obtained in this study is expected to have a good predictive ability, with satisfying predicted levels of differentiation, accuracy and practicality. The model provides a basis for clinical nursing staff to screen and timely detect breast Chinese cancer patients at risk of preoperative frailty.