Glass Health supports Epic teams that want ambient documentation and clinical decision support in the same clinician-reviewed workflow. The goal is not to replace Epic as the chart. The goal is to help clinicians move from patient conversation to note, differential, cited clinical questions, and assessment-and-plan thinking while staying grounded in the chart context they are already authorized to use.
Epic organizations usually evaluate AI through a high bar: clinical safety, clinician adoption, informatics fit, security review, and operational clarity. Glass is built for that kind of evaluation. It pairs ambient scribing with real-time clinical support, then leaves final documentation and clinical judgment with the clinician.
Why Epic teams evaluate Glass Health
Epic already has its own AI surface area, so an Epic buyer should not ask a generic question like "Can AI write notes?" The better question is more specific: what work should happen during and after the encounter, and which system should own each part?
Epic’s public AI for Clinicians page describes built-in capabilities for note summarization, AI Charting, ambient ordering, AI text assistance, inpatient insights, and other clinician workflows. That gives Epic teams a real baseline. Glass is strongest when the team wants the documentation workflow to sit next to a broader clinical reasoning layer.
For a physician, that means one working session can include:
- ambient capture of the encounter
- a draft note for review
- differential diagnosis support
- suggested history questions and exam maneuvers
- evidence-backed clinical recommendations
- cited clinical Q&A
- an assessment and plan draft that the clinician can edit before final use
That combination matters because much of the cognitive work of a visit happens before the note is final. If AI only helps after the clinician has already reasoned through the case, the tool is mainly a documentation accelerator. If it can also help organize the differential, prompt next questions, and surface evidence while the case is active, the evaluation is more clinically meaningful.
How Glass fits with Epic
Glass should be evaluated as a clinician-facing layer that supports documentation and clinical reasoning while Epic remains the system of record. In an Epic-connected workflow, authorized chart context can inform the Glass session. Glass can then help generate outputs for clinician review, including documentation and CDS artifacts.
That operating model keeps the public product boundary clean:
| Workflow area | Glass role | Epic role | Evaluation question |
|---|---|---|---|
| Encounter capture | Help turn the conversation into usable clinical documentation | Remains the chart workflow that the organization governs | Does the draft note reduce cleanup without losing clinical nuance? |
| Chart context | Use authorized context to make the session more relevant | Remains the system of record | Does chart context make the note, questions, and plan more useful? |
| Clinical reasoning | Support differential thinking, cited Q&A, and assessment-and-plan drafting | Receives final clinician-reviewed documentation through the approved workflow | Do clinicians actually use the reasoning layer during review? |
| Governance | Support a narrow pilot with clinician review and compliance review | Existing Epic, security, and operational governance remains in place | Can the team define a safe pilot before broader rollout? |
The important point is that Glass is not trying to publish an implementation specification. Clinical and operational leaders should be able to understand what the product is for, what to evaluate, and what questions to ask before rollout. Specific implementation details belong in direct review with Glass, the Epic team, security, and compliance.
What the Epic pilot should measure
A serious Epic pilot should be narrower than the buyer’s long-term ambition. Start with one service line, one or two visit types, a defined clinician cohort, and a written scorecard. That makes the review faster and reduces the chance that the pilot becomes an abstract debate about AI.
Useful pilot questions include:
- Does the note draft preserve the clinically important details of the encounter?
- Does the clinician spend less time rewriting the note?
- Does the clinician trust the chart-aware clinical support enough to use it?
- Do suggested questions, exam maneuvers, or next steps improve the live encounter?
- Do cited clinical answers reduce context switching to separate reference tools?
- Does the assessment and plan draft help the clinician move faster without weakening review?
- Does the workflow fit the organization’s Epic review and documentation expectations?
The scorecard should combine qualitative and quantitative signals. Time saved matters, but it is not the whole story. Epic teams should also review note quality, clinician burden, adoption, cleanup effort, chart-close workflow, and whether the clinical reasoning layer improves the encounter.
Stakeholder checklist for an Epic evaluation
An Epic evaluation usually fails when each stakeholder is quietly asking a different question. The physician champion may care about whether Glass helps during the visit. The informatics lead may care about whether the workflow is governable. The security team may care about PHI handling and BAA terms. Finance may care about whether there is a credible rollout path after the pilot. The pilot should make those questions explicit before the first live use.
| Stakeholder | What they should review | What a good answer looks like |
|---|---|---|
| Physician champion | Note quality, reasoning support, and whether the workflow helps during the encounter | Clinicians can point to specific cases where Glass reduced cleanup or improved the clinical review path |
| CMIO or informatics lead | Fit with Epic governance, clinician review, training, and support | The workflow is narrow enough to govern and clear enough to teach |
| Security and compliance | BAA path, production PHI terms, user access, and escalation process | The team understands who can use Glass, what is in scope, and how issues are handled |
| Operations | Pilot cohort, visit types, schedule impact, and support ownership | The rollout plan names the first users, the first visit types, and the person responsible for support |
| Finance | Plan information, pilot size, and expansion trigger | The team can connect pricing and expansion to measured workflow improvement |
Glass gives these groups a single workflow to inspect. Clinicians can evaluate the note and CDS layer. Operations can evaluate adoption and support. Security can evaluate the covered workflow. Finance can review the pricing page and decide whether a narrow pilot has a credible path to expansion. That does not eliminate diligence, but it makes the diligence more concrete.
Clinical scenarios worth testing
The best Epic pilot should include routine cases and cognitively demanding cases. If the pilot only uses simple follow-up visits, the organization may learn that Glass can draft notes, but it will not learn whether the CDS layer is useful. If the pilot only uses complex edge cases, it may miss whether the workflow is efficient enough for everyday care.
Good test scenarios include:
- a routine follow-up where the primary goal is a clean note with low cleanup time
- a new patient visit where chart context and prior history matter
- a diagnostic uncertainty visit where differential support can be evaluated
- a medication or guideline question where cited Q&A can reduce context switching
- a visit that requires a clear assessment and plan draft
- a specialty-specific note type where template fit matters
For each scenario, ask the clinician to score the output in plain language. Did the note need heavy edits? Did the suggested clinical support arrive at the right moment? Did the cited answer save time? Did the assessment and plan help organize the case? That kind of feedback is more useful than a generic satisfaction score because it shows exactly where Glass is or is not helping.
Related Epic buying guides
Epic teams comparing AI documentation tools should also read the best AI medical scribe guide and the Epic-specific best AI scribes for Epic guide. Those guides help a buyer compare the market before choosing which products deserve a deeper workflow review.
Once Epic is the environment in scope, the practical question is not just whether a product can draft documentation. The real question is whether it can support the clinician’s reasoning, review, and final documentation workflow in a way that is safe enough to pilot.
How to compare Glass with Epic AI Charting
Epic publicly describes AI Charting as a way to draft documentation, show relevant information, and queue up orders from patient conversations. That is an important built-in baseline for Epic teams.
Glass should be evaluated against a different question: do you want an ambient documentation product only, or do you want ambient documentation plus real-time clinical support in the same working session?
| Buyer question | Epic AI Charting baseline | Glass Health evaluation path |
|---|---|---|
| Can the system draft documentation from the conversation? | Epic says AI Charting can draft documentation from patient conversations | Test Glass note quality, cleanup effort, and template fit |
| Can the workflow surface relevant information? | Epic describes showing relevant information | Test whether Glass chart-aware context improves clinical usefulness |
| Can the workflow support orders? | Epic describes queueing up orders for clinician review | Confirm the current Glass workflow and final charting process directly with Glass |
| Can the clinician get clinical reasoning support during the encounter? | Evaluate Epic’s current built-in capabilities with your Epic team | Test Glass differential support, cited Q&A, suggested questions, and assessment-and-plan drafting |
| Can the team run a narrow pilot with compliance review? | Use the organization’s Epic governance process | Define the Glass pilot cohort, BAA path, clinician review flow, and success metrics |
This comparison should be practical, not theatrical. Epic AI Charting may be the right baseline for teams that want to stay fully inside Epic’s built-in AI roadmap. Glass is worth evaluating when the desired workflow includes documentation plus chart-aware clinical decision support.
Security, HIPAA, and clinician review
Healthcare AI deployment is not only a product question. For Epic organizations, the buying committee usually includes clinicians, informatics, security, compliance, operations, and finance. Glass should be reviewed with those groups from the beginning.
The public diligence path is straightforward:
- Confirm the Epic workflow with Glass and the organization’s Epic team.
- Confirm the BAA path and production PHI terms for the intended deployment.
- Define which clinicians and visit types belong in the pilot.
- Define which outputs clinicians will review before final use.
- Measure note quality, cleanup time, clinical support usefulness, and adoption.
Clinician review is not a footnote. It is the operating model. Glass can support the workflow, but the clinician remains responsible for reviewing clinical output and completing final chart actions through the approved process.
What to confirm before rollout
As a pilot moves toward rollout, Epic teams should document the exact workflow that worked. That includes the clinician-facing steps, the review path, the support model, the training plan, and the measurement plan.
The strongest rollout plans answer these questions in writing:
- Which Epic environment and clinician workflow is in scope?
- Which visit types should Glass support first?
- Which note templates or documentation types matter most?
- Which CDS outputs are expected during or after the visit?
- Who reviews output before it is finalized?
- How will clinicians report issues?
- Which metrics decide whether the pilot expands?
If the answer is still vague, keep the pilot narrow. A well-scoped Epic pilot is more valuable than a broad rollout that cannot tell whether Glass improved the work.
When Glass is a strong Epic fit
Glass is a strong fit for Epic teams that want to evaluate ambient documentation and clinical decision support together. Epic remains the system of record, and Epic-native AI is an important baseline. Glass is most relevant when the organization wants a clinician-facing layer that can support the note, the differential, cited clinical questions, and assessment-and-plan drafting in one reviewable workflow.
That evaluation should be concrete. The team should know which clinicians are in scope, which visit types are included, which outputs are reviewed, and how final documentation moves through the approved Epic workflow. If those questions are still vague, the pilot is too broad.
Good-fit Epic scenarios include:
- service lines where clinicians want documentation help plus real-time clinical support
- visits where diagnostic uncertainty, cited Q&A, or plan drafting create meaningful context switching
- workflows where chart context can make the encounter support more relevant
- teams that want to compare Glass against Epic-native AI, external scribes, and separate reference tools using observed clinician workflow value
- organizations that can align informatics, security, compliance, operations, and physician leadership around a narrow pilot
Glass may be less urgent when a team only wants to stay inside Epic’s built-in documentation roadmap. It becomes more compelling when the desired workflow includes documentation plus chart-aware CDS, differential support, and cited clinical Q&A during the same working session.
The best Epic evaluation is narrow enough to govern and specific enough to learn from. Measure note quality, cleanup effort, adoption, CDS usefulness, review flow, and support burden separately. That keeps the decision focused on real clinician workflow rather than generic AI enthusiasm.
FAQ
Does Glass Health work with Epic?
Yes. Glass supports Epic-connected workflows for ambient scribing and chart-aware clinical decision support. Teams should confirm the current supported workflow directly with Glass during evaluation.
Does Glass replace Epic AI Charting?
No. Epic remains the system of record, and Epic AI Charting is an Epic-native baseline. Glass should be evaluated when the team wants ambient documentation plus chart-aware CDS, differential support, cited Q&A, and assessment-and-plan drafting in one clinician-reviewed workflow.
What should an Epic pilot measure?
Measure note quality, cleanup effort, chart-close workflow, clinician adoption, usefulness of the CDS layer, and whether the workflow fits the organization’s Epic governance process.
Who owns final documentation?
The clinician owns final review and final chart actions through the approved workflow. Glass can help draft and organize output, but it should not be treated as a substitute for clinician judgment.
Is the Glass workflow BAA-backed?
Glass supports BAA-backed healthcare workflows. The exact terms and deployment scope should be confirmed directly with Glass before production PHI use.
What is the safest first step for an Epic organization?
Start with a narrow pilot. Pick the clinician cohort, visit type, note output, CDS output, BAA path, review model, and success metrics before rollout.
How should we compare Glass with Epic’s public AI tools?
Use Epic’s public AI for Clinicians page as the built-in baseline, then test whether Glass adds enough value through ambient CDS, differential support, cited Q&A, and assessment-and-plan drafting to justify a separate pilot.
Does Glass publish pricing?
Glass publishes plan information on the pricing page. Epic-connected deployments should still confirm current packaging, workflow scope, and contracting terms directly with Glass.