User Friendly Applications to Clinical Study Databases and Data Access

Abstract

Smith Kline & French uses a fourth generation database language because both the databases and data access programs can be modified quickly. Because we have such a powerful development tool, our clinical studies program is being developed at a rapid pace. Another tool used in support of clinical studies is a data dictionary package which fully defines all data files and data items. The primary uses of this dictionary are creating documentation and generating file descriptions of the clinical databases using a special dictionary interface. The dictionary also allows us to standardize data items and their definitions across several clinical studies. Smith Kline & French Labs is in the process of developing a standard way of collecting all clinical data items; however, these standards are not firmly defined. Our approach is to capture all the data required by the user. Since this is a new approach for us, the exact definition of what “all the data” is can never be finalized. We must allow our users to create new data items and modify existing data definitions. All requested changes are evaluated and the important ones can be incorporated into the data dictionary. These changes in the data dictionary can be passed quickly to the production databases. This entire process can be completed in a few days. Another requirement is that the data entry and quality assurance programs must be customized for each study. The user will not accept a generalized set of programs; rather, the programs for each clinical study must be designed to match the case report form. This customization serves two major purposes: The data are entered in the same form they are collected, thus reducing data entry errors; and only the data related to the clinical study are prompted for. The combined flexibility of the data dictionary and the fourth generation database language allows us to develop these customized screens in a timely manner.