Visualization of laboratory data from clinical trials

Objective: To evaluate the visual approach in the analysis of laboratory data from clinical trials.Methods: According to the real laboratory data in a clinical trials and the purpose of detecting safety signals with the requirement of drug safety evaluation in ICH,we used SAS version 9.2 and JMP/clinical3 to visualize the data.Our aim was to display the trends over time and to compare the information from two groups.Results: We illustrated here 6 specific graphical designs: boxplot over time,maximum boxplot,two-dimensional arrays of scatterplots,matrix arrays of scatterplot,dynamic bubble plot,and patient profile.Conclusion: Visualized evaluation of laboratory data can not only show summarized statistics derived from the original data but also display some original data of particular concern in laboratory data analysis.It should be primarily considered in the analysis of laboratory data arising from clinical trials.