The FDA and Duke co-founded the Clinical Trials Transformation Initiative (CTTI) to identify and promote practices that will increase the quality and efficiency of clinical trials. To this end, CTTI conceived, developed and supports AACT, a relational database that aggregates the content of ClinicalTrials.gov and makes the information freely available to the public. Our aim is to lower the bar to access this information, and challenge patients, researchers, healthcare partners and the general public to develop creative ways to make use of this important public resource.
The goal of this project is to provide the public with the aggregated content from ClinicalTrials.gov that is:
Direct Connect: AACT is a relational database in the cloud; it can be directly accessed via freely available tools such as pgAdmin, RStudio and Tableau Public, or other popular proprietary products such as SAS and Toad. For many of these tools, access is simply a matter of entering connection credentials and the database is immediately accessible to query. Click here for instructions about how to access the database directly.
Replication: AACT is a postgres database which is an open-source platform, so it's free and relatively easy to download & use locally. Postgres is well established and has a relatively low bar of entry. It is free, documentation/online support is plentiful, and the platform is light-weight and designed for this type of project. Click here for instructions about how to replicate the database.
Static Copies: AACT provides links to archived copies of the database and sets of csv files taken at monthly intervals, along with documentation for the database at the time it was copied. A static copy of the database can be downloaded, unpacked, and accessed via any number of client tools. Packages of CSV files can be downloaded, unpacked, and used as a data source for analytic tools such as SAS and R. Click here for instructions.
CTTI has attempted to make the aggregated content of ClinicalTrials.gov as intuitive and easy to interpret as possible. Each study is retrieved from the ClinicalTrials.gov API as XML, then parsed and saved as a related set of rows in the tables that make up the AACT database. The structure of the database has been designed to balance several important criteria with the aim to make it easy to understand while remaining absolutely true to the data saved in ClinicalTrials.gov. The criteria considered when designing the database and the schema design is available here
To ensure accuracy, the process that pulls data into the AACT database from ClinicalTrials.gov includes a set of validation and sanity checks.
CTTI encourages careful, creative analysis of the AACT database content and sincerely hopes people around the world will find ways to analyze this information and derive interesting, useful findings via a variety of analytical and visualization tools.