目的 构建肿瘤化疗患者发生恶心呕吐的预测模型,并验证模型的预测效果。 方法 收集 2022 年 5 月至 2024 年 5 月于 盐城市第三人民医院进行化疗的 322 例肿瘤患者的临床资料进行回顾性分析,将 2022 年 5 月至 2023 年 6 月接收的 164 例肿瘤 化疗患者为建模组,2023 年 7 月至 2024 年 5 月接收的 158 例肿瘤化疗患者为验证组,采用多因素 Logistic 回归建立肿瘤化疗患 者恶心呕吐的风险预测模型;采用受试者操作特征(ROC)曲线评估模型的预测效果;采用 Hosmer-Lemeshow(H-L)检验模型的 拟合优度。 结果 本研究 164 例患者有 92 例出现恶心呕吐,发生率为 56. 10%;结果显示,年龄、性别、抑郁、肿瘤原发部位、化疗方 案、肿瘤转移、化疗周期、卡诺夫斯基健康状况量(KPS)评分在发生和未发生恶心呕吐的肿瘤化疗患者之间存在统计学差异 (P< 0. 05);多因素 Logistic 回归分析显示,年龄(OR= 4. 555)、性别(OR= 4. 316)、化疗方案(OR= 9. 473)、肿瘤转移(OR = 30. 379)、化 疗周期(OR= 3. 823)、KPS 评分(OR= 5. 623)为肿瘤化疗患者恶心呕吐发生的影响因素(P<0. 05)。 建模组和验证组校准曲线斜 率均接近 1。 H-L 检验建模组 χ 2 = 7. 884,P= 0. 445;验证组 χ 2 = 7. 356,P= 0. 499。 ROC 曲线结果显示,建模组 ROC 的曲线下面积 为 0. 919,敏感度、特异度分别为 85. 87%、90. 28%;验证组 ROC 曲线的曲线下面积为 0. 892,敏感度、特异度分别为 90. 22%、80. 56%。 结论 年龄、性别、化疗方案、肿瘤转移、化疗周期、KPS 评分均为肿瘤化疗患者发生恶心呕吐的影响因素,构建的预测模型区分度及有效性良好,可为临床肿瘤化疗患者治疗方案的选择提供参考。

Objective To construct a predictive model for nausea and vomiting in tumor patients undergoing chemotherapy and to verify the predictive effect of the model. Method The clinical data of 322 tumor patients who received chemotherapy in Yancheng Third People's Hospital from May 2022 to May 2024 were collected for retrospective analysis. A total of 164 patients receiving chemotherapy in the hospital from May 2022 to June 2023 were selected as the modeling group and 158 patients receiving chemotherapy from July 2023 to May 2024 were selected as the validation group. Clinical data of patients were collected for retrospective analysis. Logistic regression was applied to establish the risk prediction model of nausea and vomiting in tumor chemotherapy patients receiver operating characteristic ROC curve was applied to evaluate the predictive effect of the model Hosmer-Lemeshow H-L was applied to test the goodness of fit of the model. Result In this study 92 of 164 patients had nausea and vomiting the incidence rate was 56. 10% Univariate analysis showed that age gender depression tumor primary site chemotherapy regimen tumor metastasis chemotherapy cycle and KPS score were statistically different between tumor chemotherapy patients with and without nausea and vomiting P < 0. 05 multivariate Logistic regression analysis showed that age OR = 4. 555 gender OR = 4. 316 chemotherapy regimen OR = 9. 473 tumor metastasis OR= 30. 379 chemotherapy cycle OR= 3. 823 KPS score OR= 5. 623 was a risk factor affecting the occurrence of nausea and vomiting P<0. 05 . The slopes of the calibration curves for both the modeling group and the validation group were close to 1. H-L test modeling group χ 2 = 7. 884 P = 0. 445 validation group χ 2 = 7. 356 P = 0. 499. The results of the ROC curve showed that the area under the ROC curve of the modeling group was 0. 919 and the sensitivity and specificity were 85. 87% and 90. 28% respectively the area under the ROC curve of the validation group was 0. 892 and the sensitivity and specificity were 90. 22% and 80. 56% respectively. Conclusion Age gender chemotherapy regimen tumor metastasis chemotherapy cycle and KPS score are all risk factors for nausea and vomiting in tumor chemotherapy patients. The constructed prediction model has good discrimination and validity and can provide a reference for the selection of treatment options for clinical tumor chemotherapy patients.