Application Of Repeated Measurement Data Analysis Method In Drug Efficacy Evaluation

In clinical trials,multiple measures of major efficacy indicators are required at different times,and the aggregated dataset is repeated measurement data.In the efficacy evaluation part,it is generally necessary to compare the difference between the measured values of the groups at different time points compared with the baseline,and it is necessary to verify that the effect of the test group is better than that of the control group,thereby demonstrating the effectiveness of the test study drug.In addition,it is necessary to analyze the effect of dose size and treatment time on the therapeutic effect in order to find out the best dosage and treatment cycle.Since repeated measurement data appears in various industries,the application of repeated measurement data analysis methods in China has attracted more and more attention.The analysis methods of repeated measurement data are various,and the most suitable analysis method should be selected according to the analysis target and data type.In this paper,we use several different methods of repeated measurement data to analyze the same data set,and summarize the use conditions of various methods.First,four kinds of analysis methods for repeated measurement data were used to test whether there were significant differences between the two main efficacy indicators,and whether the factors related to the treatment effect were statistically significant at each time point,and whether there is interaction between the various influencing factors.The conclusions are as follows:firstly,at the th4,8th and 12th week,the difference between the measured values and the baseline values of the two groups has significantly difference,and the measurement time was also significantly different;secondly,at the 8th and 12th week,the baseline values of the two groups have significant effect on the difference;thirdly,in the whole experimental stage,there were interactions between the baseline and the measurement time,between the baseline and the group,between the measurement time and the group.Then,an attempt was made to quantitatively analyze the changes of the two groups of measurements with time,baseline,and group.Using a mixed-effects linear model with measured values as reaction variables,group and time as fixed effects,subjects as random effects,and baseline as covariates,the predicted expressions of the two sets of measurements were obtained,further derived:when the test group and the control group had the same baseline,the test group measured higher than the control group,about 5 letters higher at the 4th week of treatment,12 letters higher at the 8th week of treatment,and about 19 letters higher at the 12th week of treatment.Finally,summarize the various methods used in the article to list their strengths and limitations.It can be seen that different methods have different requirements for data.In the analysis of actual problems,the data characteristics and research purposes should be considered simultaneously to select the most appropriate and effective method.