The Regenstrief Institute announced this week that some of its researchers have developed an eight-point framework designed to assess the validity and accuracy of algorithms to match patient medical records.
WHY IT MATTERS
Accurate patient matching across the care continuum is essential for quality and safety. It’s also key to driving down healthcare costs by reducing the ordering of duplicative medical tests. But without a national patient identifier — the United States, Regenstrief points out, is the only developed nation not to use one — that goal has been elusive for years.
Given the lack of a national patient ID, in the US, linking patient records is often dependent on algorithms designed by technology vendors or healthcare researchers.
Regenstrief vice president for data and analytics Dr. Shaun Grannis led a team of research scientists to create an eight-point framework for evaluating the performance of those patient-matching algorithms.
“Individuals increasingly receive care from multiple sources. While patient matching is complex, it is crucial to health information exchange,” Regenstrief researchers explain in announcing the framework.
“Is the William Jones seen at one healthcare system the same person as the William, Will or Willy Jones or perhaps Bill or Billy Jones receiving care at other facilities? Does Elizabeth Smith’s name appear at different medical offices or perhaps at a physical therapy or a dialysis facility as Liz or Beth? To which Juan J. Gomez do various lab test results belong?
“Typos, missing information and other data errors as well as typical variations add to the complexity,” they add.
Regenstrief’s research, supported by the Agency for Healthcare Research and Quality, has enabled a standard for matching datasets necessary for record linkage. The new framework encompasses technical areas that include data preprocessing, blocking, record adjudication, linkage evaluation and reviewer characteristics.
“Our eight-pronged approach helps to cover the waterfront of what needs to be evaluated,” said Grannis in a statement. “Laying out the framework and specifying the tasks and activities that need to be completed goes a long way toward standardizing patient matching.”
Regenstrief researchers say the framework is meant to help “provide necessary transparency” when building and validating patient matching algorithms – helping support both “internal and external validity of recording linkage studies and improving the robustness of new record linkage strategies.”
THE LARGER TREND
Efforts towards a national patient matching strategy have been ongoing for years in both the public and private sectors. There’s been some political progress toward national patient ID, but frustrations at continued roadblocks remain.
Meanwhile, research toward new matching strategies continues on several fronts. Some see biometrics as one potentially transformative approach. Other researchers have developed frameworks that achieve an accuracy as high as 99.5%.
Among other recent work at Regenstrief, Grannis and his teams have led efforts to develop HIE-trained AI models to forecast individual COVID-19 hospitalization risk, and build a population-based surveillance system using EHR data.
ON THE RECORD
“The value of data standardization is well recognized,” said Grannis in a statement about the patient matching framework. “There are national healthcare provider IDs. There are facility IDs and object identifiers. There are billing codes. There are standard vocabularies for healthcare lab test results and medical observations – such as LOINC here at Regenstrief.
“Patient identity is the last gaping hole in our health infrastructure,” he added. “We are providing a framework to evaluate patient matching algorithms for accuracy.”
Matching algorithms “come in many different flavors, shapes and sizes,” said Grannis. “To be able to compare how one performs against the other, or even to understand how they might interact together, we have to have a standard way of assessment.
“We have produced a novel, robust framework for consistent and reproducible evaluation. Simply put, the framework we’ve developed at Regenstrief provides a ‘measuring stick’ for the effectiveness of patient matching tools.”
Email the writer: email@example.com
Healthcare IT News is a HIMSS publication.