论文部分内容阅读
目的探讨随机系数模型和协方差模式模型在带有时变协变量的纵向资料分析中的应用。方法以治疗轻、中度原发性高血压病临床试验资料为例,考虑到给药方案在各个时间点随病情而变化,以用药量为时变协变量,利用随机系数模型和协方差模式模型进行分析,并通过SAS中的MIXED过程得以实现。结果两种模型拟合结果近似,组间差异无统计学意义(P﹥0.05);用药量差异有统计学意义(P﹤0.05);时间因素差异有统计学意义(P﹤0.05);年龄差异有统计学意义(P﹤0.05);治疗前舒张压差异有统计学意义(P﹤0.05)。结论随机系数模型和协方差模式模型考虑了数据相关性,考虑了时变协变量的影响,并可以处理有缺失值的资料,可以更客观的进行药物疗效评价。
Objective To explore the application of random coefficient model and covariance model in longitudinal data analysis with time-varying covariates. Methods Taking the treatment of light and moderate essential hypertension clinical trial data as an example, taking into account the dosing regimen at various time points with the disease changes, the dose is time-varying covariates, the use of random coefficient model and covariance model The model is analyzed and implemented through the MIXED process in SAS. Results The fitting results of the two models were similar, with no significant difference between the two groups (P> 0.05). There was significant difference in the dosage between the two groups (P <0.05), the difference of time was statistically significant (P <0.05) (P <0.05). There was significant difference in diastolic blood pressure before treatment (P <0.05). Conclusion The random coefficient model and covariance model consider the data correlation, consider the influence of time-varying covariates, and can handle the data with missing values, which can be more objective evaluation of drug efficacy.