Blog article

Blues patient-matching project hits 99.5{f771d91d784324d4be731abc64bffe0d1fd8f26504ceb311bcfd8e5b001778f4} accuracy

The Sequoia Project, a healthcare interoperability not-for-profit, on Thursday released a supplement to a patient-matching framework it released in 2018, this time focusing on matching people between payers.

The supplement, a case study with the Blue Cross and Blue Shield Association, reports reaching a 99.5{f771d91d784324d4be731abc64bffe0d1fd8f26504ceb311bcfd8e5b001778f4} accuracy rate when matching members to their records by creating a new member-matching algorithm, applying principles from the Sequoia Project’s 2018 framework for cross-organizational patient identity management.

Accurately matching people across care settings, payers and health information networks is foundational for achieving interoperability in healthcare, a core focus of the industry and federal government in recent years. HHS’ Office of the National Coordinator for Health Information Technology and CMS in March released companion rules on interoperability, enforcement of which has been delayed amid COVID-19.

Patient matching is “necessary in order to have interoperable health information exchange,” said Mariann Yeager, the Sequoia Project’s CEO. “You have to know that you’re talking about the same individual.”

While the Sequoia Project’s 2018 framework built upon a case study on matching patient records between healthcare providers that it completed with Intermountain Healthcare in 2016, the case study released Thursday with BCBSA adapts that framework to try to outline strategies for payers.

“BCBSA’s work around developing a person matching solution was a story we all felt needed to be told from a payer’s perspective as the focus in the industry to date has been primarily in provider settings,” a BCBSA spokesperson wrote in an email to Modern Healthcare.

Payers also face challenges when matching a member to their previous records, particularly after a member transitions to another health plan.

Accurate matching is also critical to ensuring the right health records are delivered to healthcare providers, national healthcare data networks and individual members when needed.

“Regardless of any payer’s business structure, the reality is that individuals frequently change insurance plans,” reads the case study from the Sequoia Project. “The end result for all payers is that they face the same problem of linking healthcare services provided to an individual over time.”

For its matching program, BCBSA deployed an algorithmic matching process to improve how its 36 Blue Cross and Blue Shield companies matched members’ health data between one another.

BCBSA’s proprietary algorithm assigns a unique identification number to each individual, which is used to link their records across the entire BCBS system. That number is shared with individual BCBS companies to use for tasks like requesting data from other BCBS companies that a member has previously received services from.

BCBSA’s program brings together a third-party vendor’s probabilistic matching tool—which compares multiple data elements, like members’ names and addresses, to determine the likelihood of two records being from the same member—with additional payer-specific algorithmic rules developed by BCBSA.

BCBSA also worked with a company that brings in publicly available third-party data to support and verify patient-matching decisions.

As an example of a lesson learned through the project, BCBSA shared its finding that including email addresses and phone numbers actually reduced the matching success rate for health plans, since the algorithm could get confused if a parent used their same contact information for their child. BCBSA decided those factors should not be weighed heavily in determining the likelihood of a match.

So far, the algorithm has identified more than 93.5 million separate members with 99.5{f771d91d784324d4be731abc64bffe0d1fd8f26504ceb311bcfd8e5b001778f4} matching accuracy, according to the case study.

But the matching program, while successful, involved significant workflow adjustments, training and other steps beyond developing and deploying the algorithm—the algorithm wasn’t a silver bullet.

Previously, each BCBS company had established its own data quality strategies and metrics. To successfully onboard a BCBS company to participate in the matching program, the project leaders conducted a data quality assessment and implemented changes to ensure each company had sufficient data management practices in place.

“The ability to match someone with their health data—regardless if they’ve changed insurers—is critical to ensuring people receive the care they need and deserve,” said Rich Cullen, vice president at BCBSA, in a statement. “We believe this will lay the foundation for larger health data-sharing efforts within the broader healthcare system.”


Source link