- ONC is set to launch the Patient Matching Algorithm Challenge this June in an effort to further advance interoperability.
In a recent article, Director of Standards and Technology at ONC Steve Posnack, MS, elaborated on a five minute presentation delivered alongside HIMSS Innovator in Residence at ONC Adam Culbertson regarding the upcoming challenge at Health Datapalooza 2017.
Improved patient matching has been an industry priority for years as new developments in health IT continue to spread.
The proliferation of health data exchange in particular has created a high demand for improved patient matching methods. A reliable, efficient patient matching algorithm could expedite the process of comparing different elements of a patient’s record from various health IT systems to verify the information all belongs to the same individual.
“From an interoperability perspective, the ability to complete patient matching efficiently, accurately, and at scale has long been identified as a key element of the nation’s health IT infrastructure,” wrote Posnack. “Patient matching is almost universally needed to enable the interoperability of health data for all kinds of purposes. Patient matching also requires careful consideration with respect to its effect on patient safety and administrative costs.”
A large portion of leading healthcare organizations agree optimized patient matching would significantly improve patient safety and lower administrative costs due to a reduction in duplicate patient records or information omissions caused by misplaced or mismatched patient data.
“While numerous recommendations have been issued over the years to tackle different aspects of patient matching, it is important to recognize that the entire health care system can impact its performance – from data capture at patient registration to the technology and algorithms along the way,” said Posnack. “At the same time, there has been little transparency about how well current patient matching algorithms perform and no industry-accepted minimum baselines, benchmarks or testing approaches exist.”
The federal agency’s Patient Matching Algorithm Challenge is designed to boost transparency about the performance of patient matching methods and promote the development of new algorithms.
ONC will award up to six cash prizes to the top-performing innovation developers amounting to $75,000.
Best in Category prizes will be given to participants based on best precision (fewest amount of mismatched patients), best recall (fewest amount of missed matches,) and best first F-Score run (combination of best precision and best recall.)
Innovators have up to 100 attempts to score their matching solution. The major prize category will award three prizes for the highest F-Score.
While the challenge is primarily designed to stimulate algorithm development, it will also lead to a firmer grasp across the industry of standard benchmarks and metrics for measuring patient matching performance.
Leading healthcare organizations, including CHIME and HIMSS, assert standardization is paramount in avoiding patient matching errors.
A national patient identifier assigning a unique designation to all patients across healthcare organizations could reduce error and simplify the process of comparing patients from disparate health IT systems.
HHS has yet to make the necessary adjustments to existing policy allowing for federal agencies to aid private-sector health organizations and developers in creating a national patient identifier.
ONC’s newest innovation challenge could spur useful insights to advance nationwide patient matching standardization. While a national patient identifier is still not a reality, improved standardization could mitigate pressing issues.
This challenge follows several other ONC innovation challenges designed to promote health data exchange and industry-wide interoperability.
Previous ONC innovation challenges include the Health Data Exchange Challenge promoting open application programming interface (API) innovation and the Provider Use Experience Challenge encouraging EHR accessibility through FHIR standards.