The goal of this research is to develop a new interactive software tool, called ClinTrans, to assist biostatisticians in transforming and combining clinical data in preparation for data analysis, data interchange, and cross-study pooling. Presently, the process of data preparation often requires substantial custom programming, consuming valuable resources that could otherwise be devoted to more important data exploration and interpretation activities. In Phase l, we will collect examples of a variety of clinical database layouts from the Harvard School of Public Health and from other representative academic, government, and industrial clinical research centers, and we will study the transformations that would be required to make the layouts compatible. Based on this work, we will design the detailed capabilities of the ClinTrans tool and will construct an initial software prototype to help evaluate key aspects of the design. The major technical innovation in this project is the creation of a new computer-aided approach to database transformation that is specifically tailored for handling clinical data. The major health-related contribution is hoped to be greater productivity, speed and reliability in the preparation of clinical data for analysis to support the safety and efficacy evaluation of new therapies.
