Harmonization Potential of 53 Large Population-Based Studies Using the DataSHaPER

The DataSHaPER (DataSchema and Harmonization Platform for Epidemiological Research; http://www.datashaper.org) was developed to provide a flexible, but structured approach to the harmonization and pooling of selected information between studies. In this poster/demonstration, this methodological tool is used to demonstrate the potential of sharing harmonized data (148 reference variables) between 53 large population-based studies (6.9 million participants). The DataSHaPER approach to retrospective harmonization is threefold. Firstly, rules reflecting the formal criteria that determine if a particular reference variable can be recreated from the assessment items of each study are defined. These rules also determine the quality of the match between reference variables and assessment items. Secondly, rules are applied for each reference variable and for each study participating in the harmonization process. Finally, results from this exercise are tabulated to illustrate the data sharing potential between participating studies. Results from this harmonization exercise show that a number of important reference variables can potentially be shared and co-analyzed by a large number of participating studies. The data harmonization potential thus demonstrated by the DataSHaPER tool offers the promise of greatly enhanced collaborative research generated through synthesized databases in many fields including health, environmental, and social sciences.