AI Slashes Hospice Referral Review Time from 90 Minutes to 5 as Providers Battle Intake Bottlenecks
New autonomous agent from Inferenz automates eligibility screening and admission prioritization while addressing compliance risks that plague overwhelmed hospice intake teams.

Illustration: AI-generated
Hospice providers facing referral volumes that can surge to 500 cases per day while intake teams manually process fewer than 75 now have a potential solution. Inferenz has launched a Hospice Eligibility AI Agent that reduces referral review time from 30–90 minutes to approximately 5 minutes per case, addressing what the company describes as a critical operational bottleneck in end-of-life care coordination.
The AI agent, part of Inferenz's Caregence™ agentic AI platform, became generally available on April 24, 2025, and targets a workflow challenge that creates both compliance exposure and competitive disadvantage for hospice organizations.
Automating Complex Clinical Determinations Under Time Pressure
The system processes clinical criteria required for Medicare hospice benefit eligibility, including physician-certified prognoses of six months or less, face-to-face encounter notes, Palliative Performance Scale scores, and documented clinical decline. It delivers two automated workflows: eligibility screening that returns determinations of Eligible, Pending Review, or Ineligible, and admission prioritization that ranks referrals by conversion probability.
According to Jalindar Karande, Director-Solutions at Inferenz, the technology addresses a scenario where "a typical 25-person intake team can review around 75 referrals per day, but some days bring 500 referrals." The company notes that inconsistent scoring under such pressure invites CMS clawbacks, while slower response times result in losing qualifying patients to faster-responding competitors.
The agent includes a timestamped audit trail designed for CMS review, directly addressing regulatory documentation requirements for eligibility determinations. Clinical staff can write and edit scoring rules in plain language without IT involvement, suggesting a focus on adoption by non-technical users.
Agentic AI Architecture for Healthcare Operations
Built on what Inferenz describes as Document Intelligence for data extraction and an MCP-based interoperable architecture, the system integrates with existing EHR, CRM, and intake portal systems. The company positions the technology as part of a shift toward "agentic AI" in healthcare operations—autonomous systems that execute multi-step workflows rather than single-task automation.
The Caregence™ platform includes additional AI agents for prior authorization, revenue cycle management, predictive caregiver-patient matching, and real-time risk analysis. Inferenz, based in Round Rock, Texas, describes its proprietary iDAR™ framework as enabling data-to-AI-to-ROI transformation.
Implications for Access and Compliance
For patients, the technology potentially improves access to timely hospice care by reducing the window between referral and eligibility determination—particularly important given the time-sensitive nature of end-of-life care and the six-month prognosis requirement for hospice benefit eligibility.
The system maintains clinical judgment authority over final admission decisions while reducing what the company characterizes as the cognitive burden of applying complex eligibility criteria under time pressure. The plain-language rule configuration approach suggests an implementation model that minimizes dependence on IT resources for ongoing clinical workflow adjustments.
The launch reflects broader industry movement toward AI-enabled operational automation in healthcare settings where regulatory complexity, documentation requirements, and capacity constraints intersect with time-sensitive clinical decision-making.
Sources
Reporting & accountability
How we maintain accuracy
Health AI Central separates reported facts, analysis, and public-source data. Readers can review our standards, request a correction, or consult the relevant research methodology before relying on a claim.