SpectraMedix EMPI Delivers More Than 99% Matching Engine Accuracy
- September 18, 2019
Advanced probabilistic EMPI lays the foundation for better predictive modeling
East Windsor, NJ — September 19, 2019 — SpectraMedix, a leading healthcare technology solutions provider dedicated to supporting and accelerating value-based payment and a top-10 contestant of the 2017 ONC Patient Matching Challenge, announced the release of a new, advanced EMPI (Enterprise Master Patient Index) available as part of The SpectraMedix Platform™.
“Value-Based Payment programs rely on access to clean, high-quality data,” said Raj Lakhanpal, MD, CEO of SpectraMedix. “A highly accurate EMPI is an essential resource for payers and providers, providing the whole-patient view required to facilitate value-based care. The newly advanced EMPI from SpectraMedix applies probabilistic matching for more complete, comprehensive patient matching.”
Utilizing proprietary machine learning methodologies and proven data matching algorithms, the SpectraMedix EMPI is able to probabilistically match patients with a more than 99% matching engine accuracy. This greatly improves an organization’s ability to identify high utilizers, implement risk-based models, analyze performance, and improve outcomes. Unlike deterministic matching, probabilistic matching uses a weighted measurement of demographic data and values to determine the likelihood of a match between two records. The SpectraMedix EMPI gives payers and providers a more complete, holistic perspective of a patient’s medical history across the entire health care network while helping to reduce costs by eliminating things like redundant testing.
The SpectraMedix EMPI provides the foundation necessary for accurate health care predictive models, including risk adjustment. A more complete and well-organized medical history allows for the detailed documentation and medical coding necessary for an accurate RAF (Risk Adjustment Factor) score for providers, allowing them to take advantage of an accurate capitated rate.