Regenstrief Institute won a new five-year, $1.7 million grant from the Agency for Healthcare Research and Quality for development and testing of automated patient identification approaches.
Regenstrief’s Center for Biomedical Informatics will use the funding to build on more than 15 years of work in matching patient records, officials said.
Researchers will also work with the Indiana Network for Patient Care, the largest inter-organizational clinical data repository in the country, to develop and test new approaches in a real-world patient care network.
[Also: IU, Regenstrief score $2.5 million from NIA to test new care model for patients over 50]
"Matching the correct individual to his or her health data is critical to their medical care," Shaun Grannis, MD, principal investigator for the new grant at Regenstrief, said in a statement. "Statistics show that up to one in five patient records are not accurately matched even within the same healthcare system. As many as half of patient records are mismatched when data is transferred between healthcare systems.”
Regenstrief is not the first to take on the patient ID conundrum. Industry associations including AHIMA, CHIME and HIMSS have worked toward solutions.
Recommendations on how to solve the patient identification problem have been proposed in the past, but they haven't been backed up by evidence, according to Grannis.
[Also: Health Catalyst, Regenstrief partner to commercialize natural language processing technology]
Linking patient information to the correct patient is complex for many reasons including similar "looking" patients. Many surnames are common, for example, and one in 20 Americans have the last name Smith, Johnson, Williams, Jones, Brown, Davis, Miller or Wilson.
Then, there are typographical errors, missing information, such as middle name or date of birth, and potentially identity theft. Also adding to the patient identity management problem are differing formats used by individual healthcare systems to electronically share data.
"Before you can gather clinical data, you must know exactly which patient you are talking about," he said. "Our work will help electronic medical record systems better and more accurately bring together a patient's information.”
Twitter: @Bernie_HITN
Email the writer: bernie.monegain@himssmedia.com