The Health Data Analytics Institute (HDAI) is the leader in using risk-adjusted predictive analytics that can drive measurable improvements in healthcare. Our team of executives, technologists, and data scientists have spent the last 20 years developing proprietary approaches for rapidly building and scaling rigorous risk adjustment models that can be used to assess a particular clinical approach, drug, or technique—or to quantify an individual’s risk profile—against thousands of clinical endpoints.

Our orientation is relentlessly pragmatic, taking on the minimum level of analytic complexity necessary to solve discrete use cases. By applying supervised machine learning models to existing structured data sets, we achieve results that are equivalent or superior to the most sophisticated neural net/deep learning models.  The difference is that our models are fully transparent, well calibrated across risk levels, stable within subpopulations, and easy to understand—a key requirement in facilitating clinician acceptance and uptake.  More data too often means more work and more confusion; we aim instead to synthesize and demystify, providing insights rather than data points.

Using our proprietary methodologies, we can identify which diagnoses, demographics, and procedures are most important in driving a particular outcome—and then match patients of comparable risk levels at extremely fine resolutions.  This allows us to develop highly predictive models that quantify patient risk of mortality, readmission, complications, and the development or progression of specific disease states.  We are making these predictors available through an API—available at no cost to researchers—and designing solutions for healthcare delivery, life sciences, and longevity markets.