The Centers for Medicare & Medicaid Services (CMS) publishes Comprehensive Error Rate Testing (CERT) data to highlight services with historically high billing error rates. By integrating this external dataset into AI-driven models, health payers can quickly validate whether provider outliers align with known risk areas, strengthening the evidence for preventable improper payments in prepayment or helping prioritize investigations that are most likely to yield recoveries. See below how Shift Payment Integrity maximized recoveries and future savings for this plan.
In this case, Shift’s AI applied peer group anomaly detection to identify a provider billing antigen therapy at unusually high rates compared to peers. To further validate this finding, the solution integrated external data from CMS Comprehensive Error Rate Testing (CERT), which highlights services with historically high improper payment rates. By combining internal claims analysis with external benchmarks, Shift AI not only flagged the provider as an outlier but also confirmed that the billed service was one with a high likelihood of error. This dual approach ensured accuracy in detection and enabled the health plan to take confident, corrective action at scale.
This video is part of a series of interesting cases presented by our insurance-focused team of data scientists. For more examples of Shift Technology's AI-driven results, browse the AI in Action library.