Beyond NPI: Unraveling Provider Identity Complexity

If your provider identities are “just a little wrong,” is that okay? I recently discussed managing provider identities with For the Record,  highlighting an issue that does not get the attention it deserves. But provider identities can’t be ignored – they have a profound impact on value-based reimbursement, compliance, and analytics. And as Stephanie Luthi-Terry and Jeri Romano from Allina Health shared in the article, poorly managed provider identities also affect care delivery and workflows.

Some organizations view provider identities as just a reporting issue, and if the provider identity is a little wrong, it’s still “good enough.” Organizations that think they can use the National Provider Identifier (NPI) for addressing and identifying providers are sorely mistaken, as the NPESS (National Plan and Provider Enumeration System) and PECOS (Provider Enrollment, Chain and Ownership System) data has error rates upwards of 30%, according to many current estimates. In fact, this 2013 article quotes a 48-58% error rate in the NPESS and PECOS data.

So how do you solve the provider identity challenge? Start by looking at the business problems and use cases that drive the initial creation and use of provider identities. I suggest two key points to ponder.

Consider how the use cases drive provider definition.

Identifying providers accurately and managing the associated integrity may feel a bit like managing patient identities. But I believe provider identification is even more difficult, as defining who providers are can be extremely complex. Provider data has multiple use cases, and duplication is inevitable, as providers may very legitimately have more than one NPI. A provider is much more than a physician, after all, and may include an individual (MD, PHD, MSW, PA, NP, DME, etc) an organization, a location, peripheral services that create efficiency in care (Uber partners with MedStar), and many more. This list continually evolves as healthcare transforms, so flexibility is essential.

Do you want to author and instantiate persisted provider identities, or do you want to simply index and validate the multitude of provider identifiers?

Authoring and persisting a provider identifier is difficult. You need significant exploration that involves business, clinical, and operational stakeholders, plus workflow considerations and extensive governance. Thus, many organizations start with an indexing or registry style approach to “get a handle” on provider identities. With this approach, all systems that create provider identities (particularly credentialing, application-level provider directories, and external feeds with provider data) are integrated with a registry that links the entities to a common identifier. This approach requires less governance, and offers quicker time to value and more flexibility. Many organizations or health plans/payers stay this course for years or perhaps decades.

Clearly, provider identities matter in today’s world of analytics, value-based reimbursement, and population health. Join me as I continue to explore provider identities in future blogposts.