Project Summary Background. The specific aims of the parent grant (R01 AG066793) are to conduct national epidemiological studies of Medicare and Medicaid claims to estimate the effects of long-term exposures to air pollution on Alzheimer’s disease (AD) and related dementias (ADRD) hospitalization and disease progression (Aim 1), to apply machine learning methods to identify co-occurrence of individual-level, environmental, and societal factors that lead to increase vulnerability (Aim 2), and to develop statistical methods to disentangle the effects of air pollution exposure from other confounding factors and to correct for potential outcome misclassification (Aim 3). The parent R01 relies on a wide range of epidemiological data ranging from environmental exposures, to claims data, to meteorological and socioeconomic factors. While we curated massive amounts of data for the parent R01, the data has not been deposited to a public data repository and we have not made it publicly available. Overall Goals. With this supplement our goal is to enable effective dissemination and reuse for high-dimensional high-volume data, including the data products from the parent R01. These goals will be achieved by forming a new collaboration and partnership with Harvard dataverse, expanding their current capacity to store and share geospatial public health data. Our specific aims are to: implement automatic metadata extraction for high dimensional dataset formats NetCDF and HDF5 (Aim 1), implement an integration with Jupyter Binder that allows exploration and viewing of complex high-dimensional data from Dataverse (Aim 2), enhance reuse, community engagement and reproducibility of R01 research with the demonstration of data analysis using synthetic CMS claims data (Aim 3). Impact. Each new feature in the Dataverse software platform, such as the ones proposed in Aims 1 and 2, is typically propagated to all 77 Dataverse installations, which enhances their impact worldwide. Dataverse also has a vibrant community of open-source contributors and digital libraries, and organizes annual community meetings at Harvard that last for several days. We will use this platform to promote the work in this supplement and engage the community and attend other workshops and conferences with the same goal. We will closely follow the impact of these developments through an existing Dataverse collaboration with the Make Data Count project, which provides usage metrics standardization (such as the number of views, downloads, and citations of data), and enables us to self-modify and improve our data releases. This supplement will also have a direct impact on the parent R01, enhancing reuse, community engagement and reproducibility which will in turn lead to more robust epidemiological conclusions (Aim 3).