基于HIS系统真实世界数据熄风通脑胶囊治疗急性脑梗死疗效评价研究

Evaluation of the therapeutic effect of Xifeng Tongnao Capsule on acute cerebral infarction based on HIS system real world data

关键词:  急性脑梗死  倾向性评分  熄风通脑胶囊  混杂因素  平衡

Key words:  acute cerebral infarction  propensity score  Xifeng Tongnao Capsule  confounding factor  balance effect

摘要:

[目的] 在真实临床世界，基于倾向性评分（PS）匹配法评价熄风通脑胶囊治疗急性脑梗死（ACI）的疗效。[方法] 研究数据来自于广州中医药大学附属中山中医院2021年6月—2023年6月的HIS病例资料。将其中采用熄风通脑胶囊治疗的ACI患者共240例作为观察组，并按年龄、性别资料以1∶1匹配240例未采用熄风通脑胶囊治疗的患者作为对照组。比较两组治疗有效率，分析混杂因素对处理分配的影响。采用一般化加速模型（GBM）的倾向评分平衡混杂因素。分别采用未经PS加权的Logistic回归、PS加权的Logistic回归和带协变量调整的PS加权Logistic回归3种方法分析两组患者的治疗效果。[结果] 在不考虑混杂因素的前提下，观察组的治疗有效率为68.33%，高于对照组54.17%，差异有统计学意义（P＜0.05）。采用GBM倾向评分加权法对各混杂因素进行平衡，结果显示加权前，费用类别、责任动脉、梗死面积、心脏病史、脑保护剂等协变量在两组间差异有统计学意义（P＜0.05）；加权后，费用类别、责任动脉、梗死面积、心脏病史、脑保护剂在两组间的差异无统计学意义（P＞0.05），可认为经过加权后两组协变量的分布差异基本得到平衡。分别采用未经PS加权的Logistic回归、PS加权的Logistic回归和带协变量调整的PS加权Logistic回归3种方法分析两组患者的治疗效果，结果显示无PS加权Logistic回归分的P值＜0.001，认为熄风通脑胶囊治疗ACI的疗效优于对照组；PS加权的Logistic回归P值＜0.05，认为熄风通脑胶囊治疗ACI的疗效优于对照组；带协变量的PS加权Logistic回归的P值＜0.05，认为熄风通脑胶囊治疗ACI的优于对照组。上述结果说明从统计学角度上，平衡协变量的影响之后观察组的疗效仍优于对照组。[结论] 使用熄风通脑胶囊治疗ACI比未使用熄风通脑胶囊的疗效好，经PS加权证实可提高疗效。

Abstract:

[Objective] To evaluate the efficacy of Xifeng Tongnao Capsule in the treatment of acute cerebral infarction（ACI） in the real clinical world based on propensity score（PS） matching method. [Methods] The research data comes from the data of HIS cases in Zhongshan Hospital of Traditional Chinese Medicine Affiliated to Guangzhou University of Traditional Chinese Medicine from June 2021 to June 2023. A total of 240 ACI patients treated with Xifeng Tongnao Capsules were used as the observation group，and 240 patients who were not treated with Xifeng Tongnao Capsules were matched 1∶1 according to age and gender data as the control group. The treatment efficiency of the two groups was compared，and the influence of confounding factors on treatment allocation was analyzed. The propensity score of the Generalized Boosted Model（GBM） was used to balance the confounding factors. The treatment effects of the two groups of patients were analyzed by using logistic regression without PS weighting，PS weighting Logistic regression，and PS weighting Logistic regression with covariate adjustment. [Result] Without considering confounding factors，the treatment effective rate of the observation group was 68.33%，54.17% higher than that of the control group，and the difference was statistically significant（P＜0.05）. The GBM propensity score weighting method was used to balance the various confounding factors. The results showed that before weighting，covariates such as cost category，responsible artery，infarct size，heart disease history，and cerebral protective agent were different between the two groups（P＜0.05）. After weighting，the difference in cost category，responsible artery，infarct size，heart disease history，and brain protective agent between the two groups was not significant （P＞0.05）. It can be considered that the differences in the distribution of covariates between the two groups after weighting are basically balanced. Three methods，namely logistic regression without PS weighting，logistic regression with PS weighting，and logistic regression with PS weighting with covariate adjustment，were used to analyze the treatment effect of the two groups of patients. The results showed that the P value without PS-weighted Logistic regression score was less than 0.001，it is believed that Xifeng Tongnao Capsule is better than the control group in the treatment of ACI. PS-weighted Logistic regression P value <0.05，it is considered that Xifeng Tongnao Capsule was better than the control group in the treatment of ACI. PS weighted with covariates Logistic regression was <0.05，and it is considered that Xifeng Tongnao Capsule was better than the control group in the treatment of ACI. The above results indicate that from a statistical point of view，after balancing the effects of covariates，the efficacy of the observation group was still better than that of the control group. [Conclusion] Using Xifeng Tongnao Capsules to treat ACI is better than not using Xifeng Tongnao Capsules，and it is confirmed by PS weighting to improve the efficacy.