Patient Information Quality Improvement (PIQI) Framework (Jira PSS-2607)

The generation and exchange of health data continues to grow exponentially, driven by the increasing use of data for various purposes, including the training of artificial intelligence (AI) models. Given this rapid expansion, we must address the reality that the unreliable quality of health data can lead to real patient harm. Traditional data quality criteria are no longer sufficient to meet the quality needed for modern data usage, as they often fail to address clinical complexities and advanced relationships between data elements.

As healthcare transitions towards patient-centered care, the demand for a more abstract and adaptable data quality assessment framework has intensified. Such a framework would enable healthcare organizations to evaluate data quality in a manner that aligns seamlessly with the evolving needs of patient-oriented information models, fostering greater interoperability, accuracy, and usability of healthcare data. This framework should be comprehensive and adaptable, capable of addressing the diverse needs of various health data exchange transaction formats, including at least HL7 v2 messages, HL7 CDA documents, and HL7 FHIR resources. The traditional database-centric approach often restricts the flexibility and portability required to accommodate the intricate nuances of patient-centric information models. As they are developed, we’ll need a method to computably share these advanced data quality criteria to enable seamless integration and application of the criteria across different platforms and systems, ensuring consistency and reliability in health data across the ecosystem. By establishing such a framework and associated exchange methods, we can drive data quality to higher clinical standards, reducing the risk of patient harm and improving the overall trustworthiness of health data.