The half-day before every visit
The provider network rep's job in value-based care is part account management, part operations, part clinical translator. Two or three times a quarter, they sit down with a contracted practice or medical group, walk through the numbers, and try to convince the practice manager that some workflow change is worth making before the next reconciliation closes.
The work that gets the rep into that room is mostly invisible. The night before the visit, or the morning of, the rep is pulling reports. Open HEDIS gaps from one payer portal. HCC suspecting from another. Pharmacy adherence data, if the plan has it, from a third. MAO-004 rejects for the relevant providers from a fourth. Recent ADT notifications, if the rep happens to have access. Then they assemble a PDF or a slide deck, print it, and walk in.
By the time the meeting starts, the report is already a few days old. The rep cannot answer questions the practice asks in real time, because the underlying data is back in one of the portals. The JOC, if there is one, looks at performance that closed three months ago.
The shift from explanation to action
When the prep is automated against a single canonical record, the visit changes shape. The rep walks in with the same agenda they always wanted but never had time to build: every open gap across RAF, quality, pharmacy, and care management, ranked by impact and effort, with a member-level worklist for the next 90 days.
The conversation moves from "here is what happened in Q1" to "here are the 47 members the practice can move this quarter, and here is what each one needs." The practice manager can say yes or no to specific workflow changes in the room, because the worklist is concrete enough to act on the same week.
What is actually on a Pelica-prepared practice agenda
The agenda is generated the morning of the visit, on data that is live as of that day. It includes the items the rep already wishes they had and a few they did not know to look for.
- Panel snapshot. The practice's open gaps across risk adjustment, quality, pharmacy, and care management, broken down by measure, by member, and by provider. With a current month versus prior-quarter delta so the conversation can start with what changed.
- Member-level worklist for the next 90 days. The highest-leverage opportunities, ranked by financial and clinical impact. For each member: open HCCs to recapture, due quality measures, refill timing, recent ADT events, and the recommended next action.
- MAO-004 rejection queue. Diagnoses the payer rejected, the reason code, and what documentation each one needs to be resubmitted. Most rep meetings never get to this because the data lives in a separate cycle from the rest of the work. Surfacing it on the agenda is one of the highest-RAF lifts available in any single conversation.
- Submission window timer. For each open HCC, days remaining in the relevant CMS sweep window, so the practice can sequence visits by deadline rather than by clinical convenience alone.
- Recent ADT events. Members the practice may not realize were admitted, discharged, or seen in the ED in the last few weeks. The rep can hand a follow-up list to the practice manager in the room.
- Delta versus last visit. What moved, what regressed, and what conversations from the last meeting were and were not acted on. This single piece of context turns the relationship from quarterly check-in to ongoing collaboration.
What this means for the provider, not just the plan
This is sometimes framed as a plan-side win, with the provider on the receiving end of a slicker presentation. The harder benefit is on the provider side. Practices that operate in value-based contracts already feel buried by the volume of measures, payers, and conflicting priorities. Walking out of a visit with a single ranked list of 47 specific actions, instead of seven separate reports to reconcile back to one, is the difference between a useful hour and a wasted one.
It also fixes a long-standing trust problem. Most reps cannot answer "show me this member's HCC history" without going back to a portal. When the agenda is live and the rep can pull up the underlying member record in the room, the conversation moves from "trust the report" to "look at the data with me." That shift compounds across visits.
"The hardest part of a practice visit was never the meeting. It was the four hours the night before, and walking in already knowing the data was stale."
Why this is harder than it looks
Most analytics platforms can produce a report. Getting one usable agenda in front of a rep, with current data, every time they walk into a practice, requires three things that vendor stacks rarely have together.
One canonical record per member, not one per system
Risk, quality, pharmacy, and ADT each have their own data source and update cadence. Stitching them into a single per-member view that resolves correctly across NPI, TIN, contract, and identity is a data engineering problem, not a reporting one. Without it, the rep is still reconciling rows in their head.
Submission-cycle awareness
Gap lists that ignore the CMS sweep windows produce agendas that prioritize the wrong members. Every HCC on the worklist should be tagged with days-to-window-close and submission status (in 277CA, accepted, rejected, in MAO-004). Without this, the rep cannot sequence the work.
EMR overlay, not just an outreach pop-up
When a member of the practice picks up the worklist, they need to act on it inside the EMR they already use. SMART-on-FHIR overlays into Epic, Athena, and eClinicalWorks let the agenda show up in the workflow where the documentation actually happens. Anything that lives only in a separate portal will sit there.
How the JOC changes
The biggest second-order effect is on the Joint Operating Committee. Historically the JOC is quarterly, retrospective, and ceremonial. The agenda is "here is what closed last quarter," the conversation is "we will do better next quarter," and three months later the same thing happens again.
With live data on the table, the JOC turns prospective. The agenda becomes "here is what we can move in the next four weeks, here is the worklist, here is who owns each member." The practice manager and the rep agree on three or four specific commitments, and the next JOC opens with a status update on those commitments, not with another retrospective summary. The cadence does not change. The work does.
Where to start
If you are running a provider network team that wants to test whether AI prep is worth the rest of the stack work, the most useful pilot is small and concrete.
- Pick three practices that already have a JOC cadence and where the rep is willing to try a new agenda format for one quarter.
- Define the worklist criteria: which measures, which payers, which submission windows. Keep it tight.
- Run the first agenda against live data the morning of the visit, alongside the rep's existing prep, and compare. The rep will tell you within one visit whether the agenda surfaced something they did not have.
- Track three numbers: prep time per visit, number of specific commitments made during the JOC, and how many of those commitments closed by the next visit. Those three move when the agenda is useful.