What do personalized care, customer 360 initiatives, and improving interoperability have in common? Identifying a singular patient is pivotal to their success. The systems which link together multiple patient records to one individual, called Master Patient Indexes (MPIs), have been around for decades. But the landscape surrounding MPIs has changed – patient records now traverse tens of systems and facilities or more and records are used outside of their originating organization to provide medical history and treatment details.
Traditional MPIs struggle to keep up with this pace of change because they rely on deterministic and probabilistic matching schemes, you can read more about these schemes here. These legacy identity matching schemes require a level of data standardization that cannot be guaranteed across the array of systems patient medical records traverse today.
With the increase in interoperability of systems generating and sharing patient records there are more opportunities for key data elements used to link records to be misaligned. According to the Office of the National Coordinator of Health Information, for just two facilities sharing records from the same EHR the match rate between the correct records might drop to as low as 50%. There are many reasons for low identity match rates: different facilities have their own normalization processes, data errors such as typos and alternative spellings will follow a patient for years, address updates and maiden name changes are not consistently captured, the list goes on! Patient demographic data is far from static. Unfortunately, traditional MPI methods still require a high degree of similarity before consolidating records. This misalignment causes traditional MPIs to miss relationships, overweight incorrect similarities such as a misremembered SSNs, and flag records that cannot be determined as the same person for manual investigation.
Data Stewards on Health Information Management (HIM) teams have long been the last line of defense in correcting the deficiencies of MPIs to singularly identify an individual. Data Stewards have in-depth knowledge of their patient population, organization’s data practices, and the systems responsible for their patient records. To resolve ambiguous patient identities, stewards rely on numerous tools, processes, and internal knowledge to correct these matches. A typical data stewardship task takes 10 minutes to resolve and can span many tools. But as the number of systems creating and interacting with patient records grow, so does the number of potential issues and the risk-associated with ambiguous patient identities.
Verato Auto-Steward works hand-in-hand with your data stewards using our next-generation matching engine. Among Verato’s Matching Engine’s many unique characteristics, our matching method leverages a variety of data sources such as credit header information and utility data to probabilistically match patient identity records. In a recent report from The Pew Charitable Trusts, referential matching was a key opportunity to achieving the full promise of digital health records.
Powered by Verato Referential Matching, Auto-Steward compliments a traditional MPI to automatically reduce ambiguous patient identities by 50-75%. Many of the matches Auto-Steward is able to resolve are tied to mundane data inconsistencies such as address changes and name variations.
Auto-Steward addresses a majority of the identity match questions from a traditional MPI – quickly and automatically. This empowers your data steward team to focus on resolving complicated identity match issues that are best answered by their robust knowledge and manual efforts.