Research: Patient Matching Rates Improved with Standardized Demographic Data
The consistent use of standards for demographic data would help link patients’ records from various providers regardless of where they seek care, according to new research from the Indiana University and The Pew Charitable Trusts.
Patient matching—the ability to accurately link all of a patient’s health data within and across health systems—is a problem that has plagued the healthcare industry for decades.
This latest research, published recently in the Journal of the American Medical Informatics Association, found that standardizing addresses using the U.S. Postal Service’s format—such as how to list street names and abbreviate certain words—can improve match rates by up to 3 percent. That figure can jump to as much as 8 percent when standard formats are used for both addresses and last names. If both of those details were standardized, an organization with a match rate of 84 percent, for example, could see the number of unlinked records sliced in half, the research revealed.
As such, the researchers noted that many of these types of standards are already in use in other settings, but not consistently when it comes to matching in healthcare. For this study, the research team set out to assess whether, and to what extent, the use of consistent formats for address, name, date of birth, and other demographic data yield improvements to patient matching rates.
“As medical providers make greater use of electronic health records (EHRs), improved matching will give them better access to the most complete and current health information on their patients,” wrote Ben Moscovitch, director of The Pew Charitable Trusts’ health IT initiative.
This specific study drew on tens of thousands of real-world records that had previously been matched through human review, making it possible for investigators to confirm when standardizing certain information improved the success of a matching algorithm. They found that standardizing addresses showed the most promise for improving matching, while standard formats for telephone numbers, birth dates, and Social Security numbers made no significant difference, Moscovitch said.
“Specifically, the Office of the National Coordinator for Health Information Technology (ONC) should advance the standardization of addresses. One step could be to update policies that govern how digital systems exchange information to support use of the Postal Service format. To further improve match rates, ONC should also encourage the use of additional data routinely captured in EHRs—such as email addresses—for matching,” Moscovitch said.
Included in the 21st Century Cures Act was a provision for the Government Accountability Office (GAO) to study patient record matching, and in a recent report, the GAO concluded that a lot can be done to improve patient record matching, and identified several efforts that could improve matching. To name a few: implementing common standards for recording demographic data; sharing best practices and other resources; and developing a public-private collaboration effort could each improve matching.
To this end, another recent report from the Pew Charitable Trusts outlined several key themes related to patient matching, while also suggesting recommendations to improve matching and the infrastructure needed for more robust progress in the medium and long term.
