More sophisticated technology is needed to improve patient identification matching and lessen the chance of errors, according to a report published in Perspectives in Health Information Management.

The authors of the report, led by Beth Haenke Just, MBA, RHIA, FAHIMA, analyzed almost 400,000 records that had duplicates to determine the causes of errors. "Duplicate records continue to be a significant problem in healthcare," according to the report.

Most mismatches were due to human error with a field being left blank and the provider entering in a default entry. This happened most often with either a person's middle name (58 percent of the time) or a Social Security number (53 percent of the time).

Middle names caused the most incidents of duplicate records, which in addition to blank entry errors included the middle name being swapped with the first or last name. Misspellings accounted for 53 percent of mistakes in the first name and 33 percent in the last name.

Tools such as biometrics, smart card readers and advanced algorithms are critical to reducing patient matching errors, the authors wrote, as well as enhanced search algorithms in scheduling and registration systems.

Providers can combine these new technologies with policies for name entry and standard search routines for better patient matching. 

"No amount of advanced technologies or increased data capture will completely eliminate human errors," the authors conclude. "Creating policies and procedures for front-end and back-end staff to follow is foundational for the overall data integrity process."