Conducting a clinical trial involves various stages of planning and implementation. The three major components involved in clinical trials are the management of data, the quality control to ensure data integrity, and the interpretation of the data at the conclusion of the trial. Although each process is distinct and involves different levels of effort and knowledge to implement, all processes are intimately linked. Data management techniques include the process of data entry and the implementation of an organized, comprehensive approach to quality control. Some guidelines for quality control screening are recommended to address various common issues related to clinical data, such as missing data, invalid cases, subject "outliers," and violation of distributional assumptions relevant to statistical analyses. In order to aid in interpreting the data, conditions that need to be met to make causal inferences are discussed. Taking into account baseline characteristics of the patient sample is also discussed as an extension to maintaining the internal validity of the study. Additionally, some common threats to statistical conclusion validity, including Type I error inflation and the problem of overpowered tests, are highlighted. Finally, the concept of the effect size as an important complement to statistical significance and how the various types of effect size measures can be interpreted within the context of a clinical trial are discussed.
|Number of pages||12|
|State||Published - 24 Nov 2008|
- Clinical trials
- Data interpretation
- Data management