High-frequency ED use is often framed as a “difficult patient” issue, but repeat visits usually follow predictable operational patterns: unstable medication access, inconsistent monitoring, unowned escalation decisions, and care plans that do not travel across settings. In avoidable utilization governance, the goal is not to blame people for repeated crises; it is to build control mechanisms that make stability the default. That work has to connect to primary care and care coordination, because the most effective utilization reductions come from aligned plans, shared thresholds, and follow-up that is visible and owned.
Start with a definition that is operational, not rhetorical
“Super-utilizer” programs fail when they are built on labels instead of workflows. Define the cohort using a rule that can be run monthly (or weekly) and that triggers a consistent response. Many organizations use a combination of: ED visits in a rolling window, inpatient admissions, missed primary care follow-up, behavioral health crisis contacts, and social risk markers (housing instability, caregiver collapse, food insecurity). The purpose of the definition is to activate governance—not to stigmatize.
Oversight expectations you should design for
Expectation 1: payers will expect evidence that high-cost members are managed with measurable controls. Whether under Medicaid managed care arrangements, shared savings, or quality incentive structures, organizations should anticipate scrutiny of repeat ED patterns, timeliness of follow-up, and whether care plans demonstrably change utilization trajectories.
Expectation 2: governance must be person-centered and rights-respecting while still time-bound and accountable. “Person-centered” does not mean vague. It means the plan reflects preferences and consent, while the system still sets clear safety thresholds, escalation routes, and documentation that proves decisions were made appropriately.
Build a “care plan that travels” across settings
The core asset for high-frequency ED users is a plan that is actionable in the moment of crisis. It must be accessible to the people who will use it: HCBS staff, on-call clinicians, care coordinators, and primary care teams. The plan should specify triggers, preferred interventions, contraindications, contact sequences, and what to do when the preferred option is unavailable. Most importantly, it must assign decision rights: who can authorize what, and when.
Operational example 1: Multidisciplinary case conference with decision rights and a single owned plan
What happens in day-to-day delivery. A weekly (or biweekly) case conference reviews a small, prioritized cohort identified through utilization data. Attendees include a primary care representative (or designated clinician proxy), HCBS operations lead, care coordinator, behavioral health input where relevant, and pharmacy or nurse support for medication-heavy cases. The meeting does not simply “discuss” cases; it produces a single updated plan with assigned actions, owners, deadlines, and escalation thresholds. After the conference, the coordinator publishes a one-page operational summary to frontline teams and ensures the plan is stored in the system where staff will actually look during a crisis.
Why the practice exists (failure mode it addresses). High-frequency utilization persists when responsibility is diffused. Multiple teams may be involved, but no one has the authority to align the plan and remove contradictions. The conference exists to prevent “parallel plans” that undermine each other and leave frontline staff unsure what to do.
What goes wrong if it is absent. Services become reactive: staff respond to each ED episode in isolation, the underlying drivers are never addressed, and every new clinician encounter resets decisions. Operationally, this shows up as repeated calls to 911 “to be safe,” inconsistent medication regimens, and family members receiving conflicting guidance from different parts of the system.
What observable outcome it produces. Teams can track completion of assigned actions (not just attendance), evidence that plans were updated after new events, and a reduction in repeat ED visits for members with implemented interventions. Audit trails show decisions, owners, and follow-through—making stability work visible.
Operational example 2: A crisis threshold ladder that changes what staff do before calling 911
What happens in day-to-day delivery. The plan includes a threshold ladder (green/amber/red) that defines what staff should do as risk rises: who to call first, what assessments to complete, what immediate supports to offer, and what constitutes a mandatory escalation. For example, “amber” might require a same-day clinician call and medication check; “red” might require urgent evaluation with specified alternatives to ED where available. Staff document the trigger observed, the ladder step used, and the outcome (resolved, escalated, refused, unable to contact). Supervisors review ladder usage during routine oversight to confirm it is being applied consistently.
Why the practice exists (failure mode it addresses). Without shared thresholds, escalation becomes personal preference and fear-driven risk transfer. The ladder exists to prevent “default to ED” behaviors that happen because staff lack clarity, confidence, and protection when making decisions under pressure.
What goes wrong if it is absent. Small issues become emergencies because early intervention is inconsistent. Alternatively, staff may delay escalation until deterioration is obvious, leading to worse outcomes and higher-cost care. In practice, the organization sees repeated overnight ED use, “frequent flyer” patterns that spike on weekends, and inconsistent documentation that cannot explain why escalation occurred.
What observable outcome it produces. Organizations can measure reductions in avoidable ED transports for the cohort, improved timeliness of early interventions (calls completed, visits arranged), and better documentation consistency. Reviews show fewer ambiguous escalation notes and more evidence that staff followed a defined pathway.
Operational example 3: Medication access and refill control as a utilization prevention mechanism
What happens in day-to-day delivery. For high-frequency users with medication-driven events, the plan includes a refill and access protocol: a named person monitors refill dates, verifies pharmacy pickup or delivery, and checks for prior authorization barriers. If a refill is at risk, the workflow triggers early action (contact prescriber, arrange interim supply, coordinate transportation, or use delivery services where appropriate). HCBS staff confirm on visits whether critical medications are present and taken as intended, and any discrepancy generates a same-day task for the medication owner role.
Why the practice exists (failure mode it addresses). Many repeat ED visits are the end result of predictable access breakdowns—missed refills, coverage gaps, or confusion after medication changes. This control exists to prevent “avoidable clinical deterioration” that is actually an operational failure.
What goes wrong if it is absent. People run out of medications, ration doses, or stop and restart unpredictably. Symptoms rebound, side effects escalate, and caregivers call 911 because they cannot stabilize the situation at home. The ED becomes the de facto safety net for access failures, and the system never proves where the breakdown occurred.
What observable outcome it produces. Teams can track refill reliability (on-time refill rate), reductions in medication-related ED visits for the cohort, and documented resolutions of access barriers. Audits show proactive interventions before crisis points—evidence that governance prevented events rather than merely responding afterward.
How to govern the program so it doesn’t become “extra work” that fades
Make performance visible. Maintain a small dashboard: cohort size, ED visits per 1,000 member-months for the cohort, follow-up completion timeliness, and “plan implemented” rate (actions completed as designed).
Run exception management. Every week, review members with repeat events despite a plan. The purpose is to find controllable failure modes—capacity constraints, plan not accessible, thresholds unclear, or missing decision rights.
Protect the frontline with authority. If staff are expected to reduce ED use, they must have real alternatives and clear escalation routes. Governance without operational authority simply shifts risk onto individuals.
Bottom line
High-frequency ED use improves when organizations stop treating it as a motivational problem and start treating it as a control problem. Owned plans, defined thresholds, decision rights, and auditable follow-up create stability that staff can deliver consistently—and leaders can prove under scrutiny.