Construction and validation of a risk prediction model for myelosuppression in elderly lung cancer patients undergoing chemotherapy
10.3760/cma.j.cn115682-20200831-05137
- VernacularTitle:老年肺癌化疗患者骨髓抑制风险预测模型的构建与验证
- Author:
Li CHEN
1
;
Shengqiang ZOU
;
Zhuyue JIANG
;
Jiamin HU
;
Xiaoxin YAN
;
Yaji YAO
;
Jinhan LIU
Author Information
1. 江苏大学附属镇江三院肿瘤科 212005
- Keywords:
Aged;
Lung neoplasms;
Chemotherapy;
Myelosuppression;
Risk prediction model
- From:
Chinese Journal of Modern Nursing
2021;27(14):1848-1853
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the risk factors for myelosuppression in elderly lung cancer patients undergoing chemotherapy and construct a risk prediction model for myelosuppression in elderly lung cancer patients undergoing chemotherapy.Methods:Using the convenient sampling method, data of 228 elderly patients with lung cancer undergoing chemotherapy in Respiratory Department of a Class Ⅲ Grade A hospital in Zhenjiang from May 2018 to May 2019 were selected, and risk factors of adverse reactions of myelosuppression in patients were analyzed statistically. The binomial Logistic regression was applied to construct the prediction model and the area under the ROC curve was used to test the prediction effect of the model. The patient data from January to May 2020 were collected to validate the model.Results:Among the 228 patients, 75 patients developed myelosuppression, with an incidence of 32.89%. Multivariate analysis results showed that platinum-containing chemotherapy regimens, combined with other adverse reactions, decreased albumin before chemotherapy and decreased hemoglobin before chemotherapy were independent risk factors for myelosuppression in elderly lung cancer patients during chemotherapy ( P<0.05) , which were included in the model. The area under the ROC curve of the final model was 0.823, the maximum Youden index was 0.5, sensitivity was 81.3%, and specificity was 70.5%. The results of the verification data showed that the area under the ROC curve was 0.846, sensitivity was 90.4% and specificity was 68.2%. Conclusions:The prediction effect of this model is good, which can provide reference basis for clinical treatment and formulating nursing measures to prevent myelosuppression.
