5 Steps to Close Gaps in Care and Hit Your Quality Benchmarks
A health plan flags an overdue screening and sends a care-gap alert to the listed primary care provider. The transmission succeeds—but the provider has changed practices, no one accepts responsibility, and the screening remains unscheduled.
What does “delivered” actually tell the organization?
It establishes a communication milestone, not a clinical outcome. To know whether an alert reached someone who can act, organizations need separate evidence of delivery, acknowledgment, action, and verified closure. Each stage also needs an accountable owner when progress stops.
What Is a Care-Gap Alert?
A notification about potentially missing or overdue care
A care-gap alert notifies you that recommended care may be overdue, incomplete, or undocumented in available records. It prompts a care team to review the evidence and determine the next step. The alert itself confirms neither missing care nor gap closure.
Common examples—and why they need different responses
Examples include overdue preventive screening, chronic-condition monitoring, and post-discharge follow-up. Eligibility and next steps depend on clinical context.
Abnormal-result follow-up and specialty pathway delays are related—but different—workflows. An oncology diagnostic delay may require different urgency, ownership, and completion criteria than a preventive screening reminder.
Delivered, Acknowledged, Acted On, and Closed Are Different States
Delivered: the receiving system accepted the message
Clinical message delivery confirmation establishes a defined technical event. Direct Secure Messaging notifications, interface acknowledgments, and fax confirmations have channel-specific meanings. None inherently proves clinical review.
Acknowledged: an authorized person or team accepted responsibility
Opening a message is not accepting work. Capture an explicit response: accepted, rejected with a reason, or referred for reassignment. Only acceptance establishes ownership; rejection and redirection require further follow-up.
Acted on: a documented next step occurred
Chart review, patient outreach, an order, scheduling, or documentation reconciliation can demonstrate action. Record who acted and when. Scheduling a screening is progress, not evidence that screening occurred.
Closed: evidence satisfies the applicable completion criteria
Clinical completion and quality-measure closure are separate determinations. Completed care may still require qualifying documentation. For HEDIS care gaps, use the applicable measurement-year specifications rather than a generic “completed” status.
| State | Evidence to Capture | What It Does Not Prove |
|---|---|---|
| Sent | Outbound event, destination, timestamp | Successful delivery |
| Delivered | Defined receiving-system or channel confirmation | Human review or ownership |
| Acknowledged | Explicit acceptance, rejection, or reassignment | Completed care |
| Acted on | Documented outreach, order, scheduling, or review | Gap closure |
| Closed | Completion evidence validated against applicable criteria | That the alert caused the outcome |
Why Care-Gap Alerts Stall After They Are Sent
The endpoint is valid, but the recipient is wrong
A functioning inbox can belong to an outdated practice affiliation. Payer attribution may differ from the current care relationship, and shared inboxes may lack routing rules. Endpoint validity does not establish clinical responsibility.
The alert enters a queue without an accountable owner
A queue is a destination, not an owner. Without assigned staff, coverage arrangements, and escalation rules, alerts can remain untouched during absences or organizational handoffs.
The gap reflects missing data rather than missing care
External services, delayed claims, incomplete records, and patient-matching errors can generate false-positive gaps. The absence of a claim is not conclusive evidence that care never occurred.
The next step is blocked by access or patient preferences
Scheduling capacity, transportation, language needs, and declined care require documented dispositions and appropriate support. Repeated notifications do not remove these barriers—and declined care should not automatically count as completed care.
Build a Closed-Loop Care-Gap Alert Workflow
1. Validate the patient, evidence, and gap criteria
Confirm patient matching, eligibility, exclusions, data source, and last refresh date. Identify the governing rule or measure version. Reconcile known completed care before creating another task.
2. Identify the responsible care team
Combine current provider information with care relationships and local assignment rules. A healthcare provider directory helps identify providers, locations, and endpoints; it does not independently establish responsibility for a patient.
3. Route the alert with enough context to act
Include the gap reason, supporting evidence, requested next step, urgency, sender, and reply pathway. Make the request specific enough that receiving staff can act without investigating what the sender intended.
4. Capture acknowledgment and assign ownership
Define acceptance, rejection, reassignment, and coverage rules. Record the accepting person or team. Keep rejected or redirected alerts visible until another authorized recipient accepts responsibility.
5. Track action and escalate unresolved work
Set locally approved response windows according to clinical urgency. Escalate unaccepted or inactive work to a designated owner. When automated feedback is unavailable, use documented manual follow-up rather than assuming silence means progress.
6. Validate closure and update connected systems
Return qualifying completion evidence to the originating organization. Update connected records, reconcile duplicates, and preserve exception statuses.
Before launch, confirm:
Accountable owner and validated endpoint.
Acknowledgment mechanism and escalation rule.
Closure evidence requirements.
Source-system update process.
What Data and Integrations Make Alert Tracking Possible?
Use identifiers and timestamps that connect events
Connect every event to an alert ID, patient identifier and assigning system, gap or measure ID/version, recipient, endpoint, owner, and timestamps. Preserve closure evidence references so reviewers can trace status changes to their source.
Separate message transport from workflow status
Direct Secure Messaging transports clinical content; HL7 interfaces exchange system messages; FHIR APIs exchange structured information. Their technical responses do not automatically establish human accountability.
FHIR Communication represents information exchange, Task can represent assigned work, and MeasureReport represents measure results. Implemented interfaces and agreed status definitions must connect these functions.
Bring work into existing EHR and care-management workflows
Prefer established queues and assignments where feasible. Integration requires patient matching, access controls, audit logging, appropriate information sharing, and feedback mappings. Avoid replacing disconnected spreadsheets with another disconnected alert queue.
Example: Following an Overdue Screening Alert to Resolution
When screening is genuinely overdue
Illustrative workflow—not a customer case study: A payer identifies potentially overdue colorectal cancer screening. The practice validates eligibility and available records, accepts ownership, contacts the patient, and documents the agreed plan.
Scheduling demonstrates action. After screening occurs, qualifying evidence returns to the payer for validation. Any abnormal result enters a separate follow-up workflow.
When screening was completed elsewhere
The patient reports screening at another organization. Staff obtain and reconcile external documentation, validate it against applicable criteria, and return the evidence. Resolution comes from correcting the record—not ordering unnecessary repeat care.
Measure Follow-Through, Not Just Message Volume
Track delivery, acknowledgment, and action separately
Measure:
Delivery rate by channel and destination.
Acceptance within an agreed window.
Time to first documented action.
Wrong-recipient and reassignment rates.
Unresolved-alert aging.
Manual follow-up effort.
Define denominators explicitly. Count unique alerts separately from retries, and report rejected responses separately from accepted ownership.
Calculate verified closure against a defined eligible population
Define the eligible cohort, observation period, exclusions, and evidence requirements. Calculate verified closure against eligible gaps—not messages sent. Separate newly completed care from documentation reconciliation, and avoid attributing every closure to the alert without supporting analysis.
Monitor burden and unequal follow-through
Track duplicate alerts, staff workload, and access barriers. Suppress unnecessary repetition and reserve interruptive alerts for clinically justified situations; AHRQ’s alert-fatigue guidance explains why burden matters.
Stratify follow-through appropriately to identify disparities. Communication technology can reveal barriers, but cannot resolve inequity alone.
Where careMESH Fits: Identify, Reach, and Activate the Care Team
SEARCH supports recipient identification
SEARCH provides maintained provider intelligence to support provider, location, and communication endpoint selection. Organizations still need current care relationships and assignment rules to determine who should own the alert.
CONNECT supports cross-organization communication
CONNECT provides infrastructure for delivering clinical information across organizations. Alert tracking requires clarity about the available delivery signals and what each establishes. Recipient acknowledgment and closure feedback depend on the implemented workflow and integrations.
NAVIGATE supports complex clinical-program workflows
NAVIGATE supports specialty-program coordination and patient progression across complex journeys. For oncology and other multidisciplinary programs, organizations should confirm how their configuration supports assignments, acknowledgments, and status exchange rather than assuming automatic gap detection or closure verification.
Frequently Asked Questions About Care-Gap Alerts
Does a delivery confirmation mean the provider read the alert?
No. It establishes the technical event defined by that channel and implementation—not necessarily human review, accepted ownership, or action.
What are some examples of care gaps?
Potential examples include overdue screening, missing recommended chronic-condition monitoring, and incomplete post-discharge follow-up. Each requires clinical validation and appropriate completion criteria.
Can a care-gap alert close automatically?
Yes, when an implemented process validates sufficient evidence against applicable criteria. Opening an alert or completing an outreach task is not enough.
Who owns an alert that crosses organizational boundaries?
Ownership requires explicit assignment and accepted responsibility. The sending organization should designate someone to manage failed delivery, rejection, and unresolved handoffs.
What if the receiving organization cannot return status electronically?
Use documented follow-up and evidence reconciliation. Identify who confirmed each event and when, while making delayed or incomplete visibility explicit in reporting.
Sending a care-gap alert is not the same as closing a care gap. Audit one pathway from detection through verified resolution. Find the first point where evidence or ownership disappears, then address that handoff. Reliable follow-through starts with identifying the responsible team, reaching it, and connecting communication to accountable work.