Multilevel models are applicable to both the quantitative data and categorical variables. We used the methods, including the multilevel models, analysis of covariance and CMH chi-square test, to analyse different types of data, to explore the application of multilevel models in the analysis of the multicenter clinical trial center effect. The results showed that the analysis of covariance is more sensitive to find the center effect for quantitative data, while multilevel models are more sensitive to categorical variables. It can be seen that results with different analytical methods for center effect are not the same, and the most appropriate method should be selected in accordance with the characteristics of data, the objective of research, and the applicable conditions of the various methods in practical use.
The aggregate data drug information system (ADDIS) software is a non-programming software which is based on the Bayesian framework and using the Markov chain Monte Carlo (MCMC) method for prior assessment and implementation. The operation is fairly easy for users. The consequent results and relevant plots could be output automatically by the software after users assess the consistency of model and convergence diagnostics. The major disadvantage of ADDIS is the more complicated data entry. This article introduces how to perform network meta-analysis using ADDIS software.
Objective To improve surgical results, the experience of surgical management of Ebstein anomaly in 36 cases is reported and reviewed. Methods Surgical procedures included tricuspid valve replacement (6 cases), Danielson’s operation (28 cases), Carpentier’s operation (2 cases), among them, there were 5 cases of ablation of right atrioventricular accessory pathway. Results Four patients died early after operation in the hospital, 3 from right heart failure (1 case of tricuspid valve replacement and 2 cas...
