Sentinel QA: Every Visit Note Reviewed. In Minutes. Automatically.

How Sentinel QA Works

1. A clinician submits a visit note for review, exactly as they do today. Nothing changes in the field. You choose the trigger: per visit note, per episode (SOC, ROC, recertification, discharge), by QA-queue status, on demand, or on a scheduled batch.
2. Sentinel picks the note up automatically, working through the system access your back office already has. No APIs, no integration project. Support starts with HomeCare Homebase, with more EMRs onboarding.
3. Sentinel reviews the full picture. It opens what a human reviewer would open by hand: the note plus the plan of care, physician orders, medication administration records, wound measurements, treatment flowsheets, prior documentation, and subforms. It checks completeness, internal consistency, and the compliance risks state audits surface.
4. Findings return in about 10 minutes, flagged to the clinician and the back office in the same place QA feedback lands today. No new dashboard to learn.
5. Humans stay in charge. Certified coders verify AI-drafted coding. Your QA team supervises the system, handles exceptions, and coaches clinicians on recurring issues.

From Sampling to 100% Coverage

Manual QA depends on reviewer hours, so most agencies sample: a percentage of charts, the riskiest cases, the newest clinicians. Everything else goes out the door unreviewed and becomes the chart an auditor eventually pulls. Sentinel removes the constraint. Every note gets the same review, every time, with turnaround in minutes instead of the 24 to 72 hours of a manual queue. Feedback lands while the visit is fresh in the clinician's mind, which is when documentation gets fixed.

What It Covers

• Automated visit-note QA. The core: completeness, consistency, and audit-risk review on every submitted note.
• AI-drafted coding, certified-coder verified. Coders check drafted codes instead of coding from scratch, the same standard as the rest of the Lime coding workflow.
• Per-episode and status-based QA. Reviews aligned to episode milestones and patient status changes.
• ADR support. Drafting for ADR responses from documentation Sentinel has already reviewed.
• Adjacent document drafting. High-volume back-office documents such as medical director notes for hospice organizations.

Every Finding Shows Its Work

Compliance leaders distrust black-box AI, and they should. Sentinel returns findings built for audit defense. Every finding includes:

• The issue and its severity
• The rule behind it
• The source document or field
• The exact supporting evidence from the chart
• A recommended correction

When required evidence is missing, Sentinel returns a "Cannot Evaluate" flag naming the missing source instead of guessing. Your reviewers trace each finding back to the chart. No unexplained scores, no mystery flags.

Your Rules, Not a Generic Checklist

Sentinel evaluates each chart against your organization's rule pack, spanning five families:

• Completeness and timeliness. Required fields, dates, signatures, sequencing, and documentation status.
• Clinical consistency. Visit narrative against the plan of care, physician orders, prior documentation, and changes in condition.
• Medications and treatments. Medication records, wound care, treatment parameters, and documented response.
• OASIS and episode integrity. Item logic, supporting evidence, cross-discipline consistency, and episode milestones.
• Agency policy. Your checklists, payer requirements, specialty protocols, and escalation thresholds.

The same chart reviews differently for each agency, payer, and specialty. Your QA standards, enforced on every note.

Pilots Run in Shadow Mode

You prove Sentinel on your own charts before changing anything. A pilot runs in three steps:

1. Map coverage. Confirm read-only access and chart coverage with representative test charts.
2. Shadow your QA. Load your rule pack and run Sentinel alongside your current QA process. Your reviewers grade its findings against their own.
3. Measure and expand. Track turnaround, source coverage, precision, and reviewer time. Add note types once the numbers hold.

Most agencies know within two weeks whether the findings earn trust.

Priced Per Review

You pay for QA performed. Per review, not per seat, license, or module. Cost scales with your volume, agencies start with a pilot, and we price match comparable quotes from other QA and chart-review vendors. Already using Lime's OASIS review or virtual QA staff? Sentinel extends the same coder-verified standard to every visit note automatically.