Abstract
Background We are in an era of unprecedented access to data and the potential to move areas of science forward rapidly but currently there is a lack of coordination of research efforts in regards to:
-
Using standardized measures; and
-
sharing harmonized data.
The Grid-Enabled Measures (GEM) database has been conceptualized by the National Cancer Institute (NCI) to help accomplish these goals.
Methods GEM encourages and enables standardization of measures and data harmonization by supporting a virtual community of researchers who interact with each other using Web 2.0 technologies. Through GEM, researchers are able to upload health-related measures and associated meta-data for use by others and provide comments and ratings about these measures and their associated theoretical constructs. Input from users—including ratings, usage statistics and psychometric properties of the measures—established criteria to prioritize measures to be curated for use on the Cancer Biomedical Informatics Grid (caBIG®). Curation starts with models that are used to represent the measure and its items in a form understandable to both researchers and informaticians. The outcome of the curation process is registration of a measure as a caBIG® common data element so that the variable representing the measure’s score can be shared via the grid.
Results Models for measure curation have now been developed for five measures in GEM (e.g., Center for Epidemiologic Studies Depression Scale (CESD)). GEM currently has 216 registered users and 110 measures based on 73 constructs and these numbers will change over time with user interaction. Next steps in developing GEM include making datasets—based on GEM-curated measures—available on the grid for use by other systems and users connected to caBIG®.
Conclusions GEM represents a critical step in advancing the research enterprise and complements other NIH-funded systems being developed such as Phen-X (consensus measures for Phenotypes and eXposures) which seeks to build consensus measures used in genome-wide association studies. In addition to describing the rationale and development of GEM and how measures and data are curated into caBIG®, this presentation will include a live demonstration of the site and describe the results of dissemination efforts with professional organizations.
