ABSTRACT
Data from a clinical trial will be collected and stored in some kind of computer system. The initial data capture method may be paper, a computer entry screen at the investigator’s site, a laboratory instrument, central lab system, or perhaps a hand-held computer used to collect diary information from the patient. Regardless of the way the data is captured initially, it must eventually be collected and stored in a computer system or systems that will allow complex data cleaning, reviewing, and reporting. The storage and maintenance of data is achieved by getting it into one or more databases. (A database is simply a structured set of data. This may be an Excel spreadsheet, a Microsoft Access application, an indexed proprietary data format, a collection of SAS tables, or a set of tables built in one of the traditional relational applications such as Oracle®.)
For any software that a company uses to store and manage clinical data, the data manager will design structures, a database, for each clinical trial after the protocol has been defined and the capture methods (such as case report form [CRF] booklets or electronic data capture [EDC] entry screens) have been drafted. A discussion of the design of the main capture methods can be found in Chapter 2. This chapter focuses on the design considerations that go into translating a CRF or protocol into a database for storage, cleaning, and later analysis of the data. The chapter that follows this one, Chapter 4, discusses the steps involved in putting design decisions into action: setup, building, and release of a study.
