Truveta's approach to data quality

Truveta Data is the most complete, timely and clean data on US health, with complete EHR data to study patient care with representative and precise populations. Our EHR data quality is ensured through rigorous data validation processes, including standardized data formats and robust data governance practices. 

In this whitepaper, you’ll gain insight into:

• THE TRUVETA SOLUTION – a unique system that combines artificial intelligence (AI), machine learning, and natural language processing with expert analysis. In this way, the normalization process is mostly at massive scale while leveraging clinical experts to monitor, update and evolve our system.
• SYNTACTIC NORMALIZATION PROCESS – following leading standards such as Fast Healthcare Interoperability Resources (FHIR®), the schema model is continuously added to in order to accommodate data with unique structures, while also maintaining as much uniformity across records as possible.
• SEMANTIC NORMALIZATION PROCESS – millions of normalization ‘problems’ are solved a day through a unique, machine learning model, executing at ultra-high speed with exceptional accuracy. The result is an effectively normalized data pipeline, ready for research with the very latest data from our health system members.