Urgency rises to better link patients to their records
Current patient matching efforts are marred by flaws, but unique identifiers aren’t a panacea either.
Unique patient identifiers are one of many requirements that were established by the Health Information Portability and Accountability Act of 1996 (HIPAA), which also sought standardized IDs for employer, provider and other unique identification numbers.
However, in 1998 Congress introduced language that banned the use of federal funds for the development of a UPI.
Absent a standardized national patient identifier, healthcare providers typically have created their own records numbers for patients. Matching and linking patient medical records within and across the healthcare ecosystem is done through a variety of technologies, such as enterprise-wide master patient indexes and by using records from non-healthcare sources, such as drivers’ licenses, typically called “referential matching.”
However, these techniques are not consistently effective. For instance, a Pew Report found the match rates can be as low as 80 percent - meaning one out of five patients may not be matched to all of his or her records. Even within a facility, where the patient has been seen before; the error rate can be as high as 50 percent, even when providers use the same electronic health record system. Matching then needs to be conducted manually, adding time and cost.
The increased use of patient portals has resulted in additional matching difficulties. For instance, one patient using a hospital’s patient portal inadvertently created 40 different patient records numbers for himself, says Karen Proffitt, vice president of data integrity solutions for Just Associates in Colorado.
The Health Information Technology Advisory Committee’s annual report for 2020 noted that current patient matching is imperfect, delaying care, negatively affecting interoperability and causing different patients’ information to be merged into the wrong records.
“Mis-identification is a major problem,” points out Kate McFadyen, director of government affairs for the American Health Information Management Association (AHIMA).
A UPI would not automatically perfect patient record matching. The UPI is not a substitute for having accurate patient data in the first place. There will still be typographical errors, duplicate records, data overlays, and other problems. “It’s not a silver bullet and never will be. [We still] need good processes for intake,” says Proffitt.
Gaps will also remain. For example, foreign residents or travelers won’t have a UPI.
It’s also clear that a UPI would augment, not replace current patient matching methodologies, especially in the near term. “You don’t erase linking. The UPI would be a new, important data element in how linking occurs,” says Scott Afzal, president of Audacious Inquiry, a Baltimore-based health IT company.. Nor would a UPI take the place of demographic data standardization of both formatting and collection, which are important to improve data quality and reduce errors, says Hans Buitendijk, chair of the Electronic Health Records Association in Chicago.