AI Reveals the Revenue Impact of Eligibility Accuracy

A multi-site primary care group operating under value-based care contracts in South Florida faced a quiet but costly problem: members were transferring to competing practices, and the organization was often the last to know. Under PMPM-based reimbursement models, a member lost to a competitor means lost revenue for the full month, and by the time billing cycles surface the issue, the window to act has already closed.

Sword Intelligence deployed a population-wide eligibility monitoring agent that runs continuous automated checks across all members (covering more than 4,000 unique members per month). Each week, the agent identifies members whose eligibility status has changed and surfaces actionable recovery cases for the care team.

To address the accuracy gaps in the group’s existing system, Sword Intelligence also implemented a parallel automation via the Availity web portal for Humana and CarePlus, built and deployed in under two days. The comparison was immediate: Availity completed eligibility checks for the same patient population in roughly 8 hours versus over 80 hours through the previous system, with a tech error rate below 1% versus approximately 20%. In a single week, the delta between the two systems represented over 340 undetected ineligible members: a recovery opportunity of $135,000–$270,000 that would otherwise have been missed.