Data Modeling of a Clinical Research Library

ROUSSOS, ANDRES; GUAJARDO, VIVIANA; DUBE, ERIC; BRAUN, MALENA

Berna

Congreso; 42nd Meeting of the Society for Psychotherapy Research; 2011

Society for Psychotherapy Research

The lack of a shared database is a major obstacle for the development of psychotherapy research. Around the world, clinical material has been collected and is currently being collected by independent research teams with few opportunities to share materials and methods (techniques) or to integrate existing clinical information from isolated studies. The development of a Clinical Research Library (CRL) is crucial in order to maximize the use of available clinical material. To accomplish this, it is necessary to have a system with the capacity to post, store and cross-reference clinical and quasi-clinical data sources. In the present study, the data modeling aspects of the development of a CRL will be discussed. A data model is an abstract representation of the information that will be collected and stored in a database. It identifies the information that will be contained, and the relationships among the items; it is an essential step in order to create accurate, effective and efficient databases. The data model is independent of physical constraints. A first version of a CRL data model is presented with the corresponding Entity-Relationship Diagrams (ERD). An ERD is a specialized graphic composed by entities and the relationships among them. The ERD presented in this paper will be divided in the following areas: Clinical and Quasi-Clinical Material; Analysis of the Material; participants (patient or patients, clinician, data users, among others) Subjects; Research; Case; Informed Consent; and Users. A general agreement about its usefulness, viability and value is needed to involve the research community in this demanding and worthwhile endeavor.