Agenda Day 1

Across provider networks and health plans, a similar pattern keeps surfacing: everyone senses where work breaks down, but no one can point to it end-to-end with confidence. It’s not just about dashboards; it’s about the friction points between Epic, claims, and operations. Specifically, where does the 'perfect' clinical record fall apart in the authorization process? Fragmented systems, redundancy across silos, admin waste, mismatched definitions, and delayed or duplicated feeds necessitate extra work across eligibility, authorization, discharge planning, and claims. Under cost pressure, burnout, the goal is not more dashboards, it is a shared operational picture that shows where capacity is being burned and where action actually impacts outcomes.

The industry talks about value-based care as a contract, but leaders say performance breaks in day-to-day operations. Leakage happens when patients move across pharmacy, ED, specialists, and outpatient settings without shared visibility or timely signals. The challenge is not redefining VBC, it is enabling care teams and operations to act on the same patient journey before claims arrive and margin is already lost.

Healthcare organizations are rich in data, yet critical decisions still rely on partial visibility. The reason is not missing dashboards, it’s missing context. The explanations behind denials, delayed discharges, readmissions, and patient complaints are often found in free text: physician notes, attachments, call logs, and correspondence that never make it into structured, trackable workflows. Leading organizations are now using modern analytics and AI to unlock this unstructured data and convert it into usable information, allowing teams to act earlier, improve efficiency, and focus effort where it actually matters.

As analytics, self-service, and AI expand, the biggest risk is not just missing data but the fact that decisions are made without a shared understanding. Leaders across healthcare describe the same failing model: independent teams move fast, pull data, and act on them without clarity on context or downstream impact. The issue? Not tooling or definitions alone, but the absence of visible ownership for meaning, quality, and acceptable use once data leaves the platform team. A modern stewardship layer acts less like a gatekeeper and more like a translation and accountability function, ensuring data can move quickly without undermining safety.

Pick your challenge focused seat during lunch, and get to network with peers who are focused on similar blockers - walk away with new contacts, and solutions to what’s holding you back

Interoperability has been “solved” on paper for over a decade, yet data sharing still breaks in the moments that matter most. The issue is not willingness or standards adoption, it’s usability. What arrives is often late, inconsistently formatted, or incomplete, forcing teams back into manual work just to keep care and revenue moving. It’s not an issue of data access and visibility, but rather a question of whether data can be used immediately, without rework. Progress comes from agreeing on what actually needs to move, how fresh it must be, and who owns failure when the pipeline breaks.

Readmission models are not new; almost everyone has one. The gap is between a risk score in a dashboard and a nurse or care manager doing something different before or after discharge. Whereas some are using EMR-embedded models, registries and care pathways to move from “interesting predictions”, others are implementing strategies that actually result to fewer avoidable returns to hospital.

Health plans know that moving to automated and trustable adjudication is one of the few levers left to protect margins. Claims still break out to manual review due to missing data, typos, inconsistent clinical documentation, ambiguous codes and policy exceptions. If eligibility and utilization details arrive late or incomplete, claims automation stutters, and manual work comes back. At the same time, AI promises smarter routing, fraud flags and payment accuracy, but only if the underlying data is complete and trustable.