Construction of a risk prediction model for increasing body mass index during treatment in breast cancer patients
10.3760/cma.j.cn211501-20240201-00274
- VernacularTitle:乳腺癌患者治疗期间体质量指数增长风险预测模型的构建
- Author:
Qing WANG
1
;
Xinjie JIA
;
Xue WU
;
Miao WANG
;
Xin HE
Author Information
1. 天津医科大学肿瘤医院乳腺二科 国家恶性肿瘤临床医学研究中心 天津市肿瘤防治重点实验室 天津市恶性肿瘤临床医学研究中心 乳腺癌防治教育部重点实验室，天津 300060
- Keywords:
breast neoplasms;
body mass index;
risk factors;
prediction
- From:
Chinese Journal of Practical Nursing
2024;40(31):2422-2429
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate the independent influencing factors of body mass index (BMI) growth risk during breast cancer treatment and establish a predictive model, and validate its predictive efficacy.Methods:A retrospective cohort study was conducted by a convenience sampling method. 306 eligible breast cancer patients underwent follow-up at the Tianjin Medical University Cancer Hospital between January and May 2023 were selected as the modeling group. The BMI changes of the modeling group patients were monitored during the study period. Patients with BMI changed < ± 0.5 kg/m 2 were classified as the stable group, while those with BMI changed ≥ 0.5 kg/m 2 were classified as the growth group. Factor analysis was conducted, and a predictive model was established. The fitting effect was verified by the Hosmer-Lemeshow test, and the receiver operating characteristic (ROC) curve was drawn to verify the model′s predictive ability. 110 eligible breast cancer patients underwent follow-up at the Tianjin Medical University Cancer Hospital between June and August 2023 were selected as the validation group to verify the predictive effect of the model. Results:The age of the modeling group patients was (48.89 ± 7.78) years, BMI was (25.53 ± 1.39) kg/m 2 and the age of the validation group patients was (50.18 ± 7.70) years, BMI was(24.31 ± 0.86) kg/m 2, both groups were female, there were no significant differences between the two groups (all P>0.05). Axillary lymph node dissection ( OR=4.196, 95% CI 3.512-9.158), chemotherapy duration ( OR=2.002, 95% CI 1.001-3.003), premenopausal status ( OR=3.257, 95% CI 1.244-3.733), low health behavior capacity ( OR=5.977, 95% CI 2.878-7.893), and low physical activity ( OR=3.755, 95% CI 1.244-11.733) were identified as independent influencing factors of BMI change during breast cancer treatment (all P<0.05). The predictive model was P=0.915, with the ROC curve area of 0.950, a J-index of 0.785, a sensitivity of 0.971, and a specificity of 0.814. Internal and external validation of the model in the validation group revealed a sensitivity of 0.858, a specificity of 0.887, and an accuracy of 97.3%. Conclusion:The predictive model exhibited excellent performance and can serve as a valuable tool for formulating nursing intervention strategies to maintain the BMI stability of breast cancer patients during treatment.
