Skip to main content
Call

For wound-center and hyperbaric service-line leaders, patient-access and prior-authorization teams, clinical documentation and health-information leaders, product and supply operations, compliance, coding, charge-capture, and revenue-cycle leaders who need a defensible administrative record without asking software to make clinical, coverage, coding, or product-selection decisions.

Turn Every Wound Episode Into a Review-Ready Evidence Record

Create one versioned administrative record for the wound episode under review: wound identity, clinician-authored measurements and descriptions, dated photographs as governed artifacts, conservative-care history, vascular and infection context, offloading or compression records, debridement documentation, skin-substitute request and product ledger, hyperbaric oxygen request, payer instructions, submission history, response scope, and unresolved questions. Every fact keeps its source, date, author, and human approval state.

This is provider-operations software, not a clinician, coding authority, utilization reviewer, inventory system of record, or payer. It does not examine a photograph, calculate a wound measurement, diagnose or stage a wound, decide that conservative care failed, determine infection or vascular status, recommend debridement, skin-substitute use, a product, an application count, or hyperbaric oxygen therapy, establish medical necessity, select codes or modifiers, submit an attestation, or promise authorization or payment. Qualified humans retain every clinical, coverage, coding, procurement, financial, compliance, privacy, and legal decision.

The record can look complete while the episode is still unprovable

Wound care produces evidence over time and across teams. A current note cannot repair a missing baseline; a photograph cannot explain itself; a product invoice cannot prove what was applied; and an authorization headline cannot establish that the approved service, site, dates, product, and later claim still match. Staff often discover those gaps only when a request, additional-documentation review, claim edit, or appeal is already waiting.

  1. Serial evidence has no dependable episode spine

    Measurements, stage or depth wording, treatment descriptions, photographs, vascular records, infection statements, and procedure notes may live in different modules or arrive from outside facilities. Dates can be clear while wound identity, laterality, anatomical wording, author, or amendment state is not. Copy-forward text can make an older observation look current.

    Operational consequence

    Authorization and revenue-cycle staff spend time rebuilding chronology, yet a reviewer can still receive an unsupported gap, a record from the wrong wound, or a later entry presented as if it were the baseline. The case cycles back to the clinical team and the submission clock keeps moving.

  2. Conservative-care facts are present but not safely interpretable

    The chart may mention dressings, offloading, compression, debridement, vascular workup, infection management, or another intervention without an explicit clinician-authored duration, adherence statement, response, contraindication, or reason for change. Administrative staff then face pressure to translate activity into a clinical conclusion that the source does not make.

    Operational consequence

    A request can overstate what the record supports or remain idle while staff repeatedly ask broad questions. Both outcomes consume clinician time, and neither creates a reusable, source-linked explanation of what was documented and what still needs qualified judgment.

  3. Photographs and measurements travel without their boundaries

    A thumbnail, exported image, ruler photograph, wound diagram, device-generated measurement, or scanned page may lose its original link, capture context, rendition, or consent and disclosure status. A numeric value can also conflict with the signed note or use a different method without explanation.

    Operational consequence

    Teams risk attaching the wrong artifact, disclosing more than the request requires, or implying that software validated clinical content. Privacy, records, and clinician review arrive late, and an otherwise usable packet returns to a hold state.

  4. The product, procedure, and wastage story splits after approval

    A skin-substitute request, authorization response, purchase record, packing reference, lot or batch identifier, procedure note, charge, units used, units discarded, and reason for wastage may be owned by different teams. Debridement depth and procedure count can also diverge between the signed note and the administrative ledger.

    Operational consequence

    Supply, coding, charge-capture, compliance, and billing teams reconcile the episode after the fact. A product or quantity mismatch may remain hidden until claim review, and correcting it can require several owners to reconstruct who knew what at each step.

  5. Coverage context changes faster than the packet

    A payer product, Medicare jurisdiction, model participation, plan policy, portal instruction, site, rendering provider, product, planned service, or date can change after the first review. Skin-substitute coverage and hyperbaric oxygen requirements do not come from one nationwide checklist, and one payer's policy cannot be reused for another.

    Operational consequence

    A green work-queue label can outlive the assumptions behind it. Scheduling, product acquisition, treatment, or billing may proceed against the wrong scope unless the workflow preserves the prior version and reopens the exact human decisions affected by the change.

A named agent team with visible decision boundaries

Each agent handles a defined part of the longitudinal wound evidence and skin-substitute compliance workflow. Confidence signals and human escalation remain visible rather than being hidden behind an automation label.

  1. Wound Episode Identity Agent

    This AI agent opens and maintains the administrative identity of one wound episode and one requested workflow. It links incoming records without merging distinct wounds, services, sites, or payer contexts, and it preserves earlier versions so staff can see how the request changed over time.

    Inputs

    • Buyer-approved scheduling, referral, registration, and order feeds
    • Signed wound notes and document metadata from approved systems
    • Coverage, payer-product, provider, facility, and service context
    • Prior packet, response, and claim references approved for reuse review

    Checks

    • Wound reference, anatomical wording, side, episode dates, and source agreement
    • Requested service, provider, facility, payer product, and work-type alignment
    • Duplicate episode, copied-forward content, amendment, and superseded-version risk
    • Whether a change should reopen only affected checks or create a new episode

    Outputs

    • A versioned episode header with source provenance
    • A lineage map across prior requests, responses, and corrections
    • An owner-routed queue for identity conflicts and missing scope
    Confidence method
    Confidence reflects exact agreement across approved identifiers, document metadata, dates, wound wording, and current request scope. It describes linkage reliability only; it never represents clinical certainty, wound severity, medical necessity, coverage, or readiness to treat.
    Low-confidence action
    If records could belong to different wounds, the side or site conflicts, a copied note appears newer than its source, or request scope is ambiguous, the agent keeps the records separate and blocks downstream assembly. It never resolves identity by guessing from nearby text.
    Human escalation
    Patient-access or records staff resolve administrative identity; the treating clinician confirms clinical wound identity and intent; authorization and revenue-cycle owners confirm payer and service scope. Privacy, compliance, or legal leaders decide disputed linkage or disclosure questions.
  2. Serial Wound Evidence Librarian

    This AI agent builds a source-linked chronology of clinician-authored wound descriptions, measurements, stage or depth wording, vascular and infection references, and documented response. It indexes what the record says without calculating a trend, assigning a stage, or deciding whether progress is clinically sufficient.

    Inputs

    • Signed progress notes, procedure notes, and approved flowsheet exports
    • Source-system measurement values and method metadata
    • Clinician-authored vascular, infection, and wound-response statements
    • Current payer question set or documentation request

    Checks

    • Date, author, wound identity, measurement units, and method provenance
    • Baseline, interim, pre-procedure, and post-procedure source availability
    • Conflicts between narrative, flowsheet, device export, and later amendment
    • Unsupported trend language, missing context, and evidence copied from another episode

    Outputs

    • A dated wound-evidence timeline with direct source links
    • A comparison view that preserves conflicting values and methods
    • A clarification queue aligned to the exact unanswered question
    Confidence method
    Confidence is based on authenticated source, stable wound linkage, explicit authorship, unit and method context, and chronological consistency. A complete-looking series does not increase confidence in clinical meaning when the signed record does not supply that meaning.
    Low-confidence action
    When a value lacks provenance, units or method differ, a stage or depth conflicts, or a response statement is absent, the agent labels the issue and asks for qualified review. It does not calculate area reduction, infer healing, normalize measurements, or choose the favorable entry.
    Human escalation
    The treating wound clinician validates clinical descriptions, measurements, staging, depth, response, infection, and vascular meaning. Health-information staff authenticate records, and authorization or revenue-cycle staff use only clinician-approved statements in payer or claim workflows.
  3. Conservative-Care Qualification Mapper

    This AI agent organizes the documented history of dressings, offloading, compression, debridement, vascular assessment, infection management, and other conservative care against a current, human-selected question set. It distinguishes a recorded event from a clinician's conclusion about adherence, response, failure, or contraindication.

    Inputs

    • Signed treatment plans, progress notes, orders, and procedure documentation
    • Approved device, supply, home-health, referral, or outside-record evidence
    • Human-selected current payer policy, form, or review question
    • Clinician-approved clarifications created for the active request

    Checks

    • Therapy name, duration wording, source date, owner, and wound linkage
    • Whether adherence, intolerance, contraindication, response, or failure is explicit
    • Presence and provenance of vascular assessment and infection-status statements
    • Policy-version, jurisdiction, payer-product, and request-scope match

    Outputs

    • A conservative-care chronology separated by evidence category
    • A question-to-source map with unsupported conclusions exposed
    • Targeted tasks for clinician, records, supply, or authorization owners
    Confidence method
    Confidence measures source completeness and exact responsiveness to the selected administrative question. Orders, supply delivery, device presence, or recurring note text cannot by themselves establish use, adherence, clinical response, failed care, or qualification for a requested service.
    Low-confidence action
    If duration is unclear, evidence comes from an unverified outside record, the policy question changed, or only activity rather than clinical meaning is documented, the agent holds the mapped answer. It never converts frequency, possession, or elapsed time into clinical qualification.
    Human escalation
    The treating clinician authors or approves all clinical conclusions. Authorization staff confirm the current payer question; records and supply owners verify provenance; compliance and coding professionals address documentation or billing risk. The agent never signs a qualification statement.
  4. Clinical Media Custodian

    This AI agent manages the administrative custody of serial wound photographs, diagrams, scans, and device exports. It identifies artifact type, origin, linkage, rendition, amendment, and approved disclosure status while deliberately avoiding pixel analysis or any visual clinical interpretation.

    Inputs

    • Buyer-approved media manifests and note-to-artifact links
    • Capture time, uploader or author, source device, rendition, and version metadata
    • Consent, minimum-necessary, redaction, retention, and export rules
    • Current payer attachment request and clinician-approved artifact set

    Checks

    • Artifact origin, wound and encounter linkage, timestamp, and version lineage
    • Original versus thumbnail, cropped copy, annotated derivative, or unsupported format
    • Whether a narrative, photograph, ruler image, diagram, or device export was requested
    • Recipient, consent, redaction, sensitive metadata, and disclosure authorization

    Outputs

    • A non-diagnostic media inventory with provenance and use restrictions
    • An attachment manifest separating approved and excluded artifacts
    • A clinician, records, privacy, or security review queue
    Confidence method
    Confidence describes artifact identity and custody, not the wound shown. It depends on authenticated origin, reliable note linkage, consistent file lineage, readable administrative metadata, and a matching buyer rule for disclosure, retention, and export.
    Low-confidence action
    If origin, wound linkage, rendition, consent, redaction, recipient, or requested artifact type is uncertain, the agent blocks attachment and preserves the original reference. It never measures pixels, compares appearance, detects tissue, assigns stage, or infers improvement or deterioration.
    Human escalation
    A qualified wound clinician interprets every image and approves any clinical statement. Health-information, privacy, and security staff control custody, consent, redaction, disclosure, and retention. Authorization staff attach only the specifically approved artifact set.
  5. Product and Procedure Ledger Agent

    This AI agent reconciles administrative facts across a planned or completed skin-substitute application, debridement, or related wound procedure. It connects the human-approved request, product record, lot or batch reference, quantities used and discarded, reason for wastage, procedure note, charge, and claim draft without selecting a product or code.

    Inputs

    • Human-approved procedure plan and current authorization scope
    • Buyer-approved inventory, procurement, packing, lot, batch, and receipt records
    • Signed procedure note and clinician-authored debridement or application details
    • Charge-capture, coding-review, and claim-draft records

    Checks

    • Product name, identifier, lot or batch, source, and date agreement
    • Requested, received, used, discarded, returned, and charged quantity consistency
    • Documented reason for wastage and best-fit review evidence when required
    • Procedure count, clinician-authored depth, service date, site, and claim-scope mismatch

    Outputs

    • A source-linked product and procedure ledger
    • A pre-bill mismatch queue for supply, clinician, coding, and charge owners
    • A preserved amendment trail rather than an overwritten transaction
    Confidence method
    Confidence reflects reconciliation across authenticated administrative and signed clinical sources. Matching totals do not prove that a product choice, application, debridement depth, unit, code, modifier, charge, or claim is clinically appropriate, covered, or payable.
    Low-confidence action
    If lot lineage is incomplete, quantities do not reconcile, the procedure note conflicts with inventory, or the coding basis is unclear, the agent holds downstream release and names the mismatch. It never invents wastage, edits a signed note, or selects a product, unit, code, or modifier.
    Human escalation
    Supply and inventory owners validate product custody; the treating clinician validates the performed service and clinical details; qualified coding and charge-capture professionals approve units and claims; compliance, finance, and legal leaders handle unresolved product, wastage, or billing risk.
  6. Skin-Substitute and HBOT Packet Builder

    This AI agent assembles a draft skin-substitute, cellular or tissue product, or hyperbaric oxygen administrative packet from human-approved evidence and current payer instructions. It shows missing answers and source versions rather than hiding them inside a generated narrative.

    Inputs

    • Confirmed episode, payer product, benefit lane, provider, site, and requested service
    • Human-approved serial evidence and conservative-care maps
    • Current payer forms, portals, policies, LCD or NCD references, and attachment rules
    • Approved signatures, attestations, product context, and submission controls

    Checks

    • Question-to-source traceability and attachment-version completeness
    • Payer, jurisdiction, plan, service, site, provider, product, and date agreement
    • Separate skin-substitute, debridement, and HBOT evidence requirements
    • Missing clinical decision, signature, consent, portal field, or duplicate-submission risk

    Outputs

    • A draft packet with evidence manifest and explicit open items
    • Separate clinical, administrative, coding, and submission approval gates
    • A submission-ready package only after configured human approvals
    Confidence method
    Confidence measures completeness against the current human-selected question set and approved sources, not the probability of authorization or payment. The packet exposes the policy or instruction version, applicable scope, unanswered items, and channel limitations.
    Low-confidence action
    If a requirement, jurisdiction, service category, policy version, attachment, or signature is uncertain, the agent keeps the packet in draft and routes the question. It does not choose the most favorable policy, fabricate clinical support, attest, or submit through an unvalidated channel.
    Human escalation
    Authorization staff approve administrative completeness; the treating clinician approves clinical statements and attestations; coding, compliance, privacy, and supply staff approve their domains; and an authorized submitter releases the packet. Payers retain coverage authority.
  7. Response and Change-Control Coordinator

    This AI agent links payer acknowledgements and responses to the exact submitted packet, then opens accountable work for information requests, partial scope, denials, appeals, renewals, service changes, product changes, and post-procedure reconciliation. It keeps an earlier decision immutable when facts change.

    Inputs

    • Authenticated portal, transaction, fax, letter, and call-reference records
    • Submitted packet version and evidence manifest
    • Current schedule, service, site, provider, product, and payer context
    • Buyer-approved response, appeal, renewal, release, and fallback rules

    Checks

    • Response identity, date, status, conditions, expiration, and packet linkage
    • Approved versus current service, product, provider, site, quantity basis, and dates
    • Information-request, denial, appeal, reconsideration, and deadline instructions
    • Downstream acknowledgement and any change that invalidates an earlier checkpoint

    Outputs

    • A response-scope comparison with conditions and mismatches visible
    • Owner-routed proceed, hold, correct, resubmit, appeal, or renewal tasks
    • A complete version lineage through closure and claim follow-through
    Confidence method
    Confidence depends on authenticated response source, exact linkage to the submitted version, and agreement across configured scope fields. A portal headline, old authorization number, or verbal reference without reliable linkage cannot establish current operational readiness.
    Low-confidence action
    When response scope, deadline, product, service, site, provider, or appeal level is uncertain, the agent creates an urgent human review and preserves competing evidence. It never infers approval, extends dates, releases treatment or product, files an appeal, or promises payment.
    Human escalation
    Patient-access and authorization staff validate responses; clinicians control treatment decisions; supply and scheduling teams control product and appointment release; revenue-cycle staff control financial clearance and claim action; payer, compliance, and legal experts resolve disputes and deadlines.

The operating sequence, evidence by evidence

The sequence separates agent work from accountable human checkpoints so teams can see what moves forward, what stays pending, and why.

  1. Wound-center intake and patient access

    1. Define the wound episode and the work being requested

    The workflow begins with one explicit administrative question: prepare a skin-substitute request, build an HBOT packet, answer an information request, reconcile a procedure, support a claim review, or organize an appeal. The episode header records source, wound reference, provider, site, payer product, service, and version while keeping prior work linked but unchanged.

    Agent actions

    • Create the episode header from approved sources
    • Separate distinct wounds, services, payer products, and work types
    • Link prior packets and responses without copying their conclusions
    • Route identity, ownership, and duplicate-work conflicts

    Evidence produced

    • Versioned episode and request identifier
    • Source and lineage map for prior work
    • Scope exceptions and accountable owner queue

    Human checkpoint: Intake staff confirm administrative identity and ownership; the treating clinician confirms the wound and intended clinical service; authorization staff confirm the active payer and work type. No clinical urgency, qualification, or coverage conclusion is created at intake.

  2. Clinical documentation and health information

    2. Build the longitudinal evidence map

    The Serial Wound Evidence Librarian and Clinical Media Custodian index signed descriptions, measurements, procedures, and governed artifacts in chronological order. The map preserves method, author, amendment, and conflicts so a later note or photograph cannot silently stand in for a missing baseline.

    Agent actions

    • Index signed wound descriptions and source-system values
    • Create non-diagnostic links to approved photographs and diagrams
    • Expose missing baseline, interval, pre-service, or post-service evidence
    • Hold conflicts for clinical and records review

    Evidence produced

    • Source-linked timeline and artifact manifest
    • Measurement and wording conflict report
    • Missing-evidence tasks tied to exact questions

    Human checkpoint: A qualified wound clinician validates clinical meaning and any comparison; records and privacy staff validate provenance and disclosure. The agent cannot measure, stage, interpret, or decide whether the wound improved or worsened.

  3. Authorization coordinator with clinician review

    3. Map conservative care and qualification questions

    The Conservative-Care Qualification Mapper organizes documented offloading, compression, dressings, debridement, vascular assessment, infection management, and related care against the current question set. It marks the difference between an event and an explicit clinician-authored statement about adherence, response, failure, or contraindication.

    Agent actions

    • Build category-specific treatment chronologies
    • Map each current question to an approved source
    • Identify unsupported clinical conclusions and stale evidence
    • Create narrow clarification tasks instead of broad chart requests

    Evidence produced

    • Conservative-care evidence matrix
    • Clinician-approved clarification record
    • Open qualification questions with policy-version context

    Human checkpoint: The treating clinician supplies every clinical conclusion and decides whether the documented history supports the proposed care. Authorization staff confirm that approved evidence answers the current payer question but do not convert document presence into medical necessity.

  4. Patient access, reimbursement, and compliance

    4. Verify the controlling payer path and current instructions

    The team selects Original Medicare, Medicare Advantage, Medicaid fee for service or managed care, commercial, or liability workflow and records the controlling jurisdiction, plan, policy, form, portal, and effective date. Current official sources take priority over an old packet, training slide, or generalized checklist.

    Agent actions

    • Separate program, plan, jurisdiction, and benefit context
    • Record authoritative source and effective-date provenance
    • Compare the current service and product with the selected instructions
    • Route ambiguous or conflicting policy context to a qualified human

    Evidence produced

    • Human-approved payer-route record
    • Current policy and instruction reference set
    • Coverage-context conflicts and manual-verification log

    Human checkpoint: Authorized reimbursement, contracting, compliance, or payer-relations staff choose the administrative route and confirm the current source. The payer makes coverage determinations; clinicians make care decisions; legal counsel handles disputed rights or jurisdictional questions.

  5. Authorization and supply operations

    5. Assemble the request and reconcile product dependencies

    The Packet Builder composes the draft question-to-source package while the Product and Procedure Ledger Agent checks any approved product, lot, quantity, wastage, procedure, or charge context already in scope. Skin-substitute and HBOT work remain distinct even when they draw from the same wound chronology.

    Agent actions

    • Assemble only human-approved answers and attachments
    • Reconcile proposed request scope with product and service context
    • Expose missing signatures, attestations, and domain approvals
    • Prepare a submission manifest and tested manual fallback

    Evidence produced

    • Draft packet and attachment manifest
    • Product or procedure dependency report
    • Clinical, administrative, privacy, coding, and submission approval state

    Human checkpoint: The clinician approves clinical statements and the proposed care; supply, coding, privacy, compliance, and authorization owners approve their respective facts. An authorized submitter must approve release, and unresolved items remain visible rather than being generated away.

  6. Authorized submitter and patient access

    6. Submit, reconcile the response, and control release

    After required approvals, the exact packet version is released through a validated channel and its acknowledgement is recorded. The response is linked back to that version and compared with current service, product, site, provider, dates, conditions, and downstream readiness controls.

    Agent actions

    • Record submission version, channel, user, time, and acknowledgement
    • Authenticate and classify the response without inferring scope
    • Compare response terms with the current operational plan
    • Route information, correction, denial, appeal, or hold work

    Evidence produced

    • Submission and acknowledgement record
    • Response-scope reconciliation
    • Named downstream handoffs and unresolved conditions

    Human checkpoint: Patient-access staff validate the response; clinicians control treatment; scheduling and supply teams apply release rules; revenue-cycle staff confirm financial readiness. Authorization does not guarantee coverage or payment, and no automated status bypasses a required human owner.

  7. Revenue cycle, compliance, and service-line operations

    7. Reopen changes and close with product-to-claim lineage

    A product, procedure, schedule, site, provider, payer, response, or treatment-course change opens a new version and rechecks only affected decisions. After a service, the ledger compares approved scope, signed documentation, product custody, wastage, charges, and claim draft before accountable humans close the episode.

    Agent actions

    • Detect approved change events and preserve the prior version
    • Reopen affected authorization, clinical, supply, or coding checks
    • Reconcile signed procedure and product records with charge context
    • Record closure reason, exceptions, acknowledgements, and retention status

    Evidence produced

    • Immutable change and approval history
    • Pre-bill product, procedure, and charge reconciliation
    • Closed episode record with remaining exceptions named

    Human checkpoint: Clinical, supply, coding, charge-capture, billing, compliance, and legal owners resolve their exceptions and authorize closure. The agent cannot edit signed documentation, decide a code, release a claim, waive a balance, or determine audit or appeal strategy.

Separate payer paths instead of one universal rule

Coverage, notice, authorization, and documentation requirements vary by program, plan, jurisdiction, service, and date. These paths show where teams must verify current authoritative instructions.

medicare

Original Medicare: jurisdiction, service, and date come first

Original Medicare wound workflows can involve national coverage, MAC-specific LCDs and billing articles, claim-by-claim review, and—in selected states and services—the WISeR model. CMS withdrew certain new skin-substitute LCDs that had been scheduled for January 1, 2026, while pre-existing local policies may remain active. The correct current source must be verified for the service location and date; no single skin-substitute checklist is nationwide.

  • Verify the current MAC, jurisdiction, effective LCD or article version, and exact service rather than relying on a neighboring jurisdiction or archived packet.
  • Treat CMS's WISeR model as limited to Original Medicare, selected states, and selected services; the model does not itself change Medicare coverage or payment policy.
  • Handle hyperbaric oxygen under the applicable national and local sources. CMS NCD 20.29 includes condition-specific criteria and continuing-evaluation language, but the agent cannot decide whether a real case meets them.
  • Preserve baseline, serial, conservative-care, product, procedure, and response evidence requested by the controlling source without implying that document presence establishes medical necessity.

Human handoff: Medicare reimbursement and compliance staff verify the current MCD sources, MAC instructions, WISeR applicability, submission or prepayment-review route, and coding guidance. The treating clinician supplies clinical meaning; an authorized submitter releases the request; the MAC or model review process retains determination authority.

Sources for this path: Centers for Medicare & Medicaid Services, Centers for Medicare & Medicaid Services Medicare Coverage Database, Centers for Medicare & Medicaid Services Innovation Center, Centers for Medicare & Medicaid Services Medicare Coverage Database

medicare advantage

Medicare Advantage: base-benefit rules plus the current plan

A Medicare Advantage plan is not the Original Medicare WISeR path. CMS rules address how MA organizations use Traditional Medicare coverage criteria and when internal coverage criteria may be used, but the team still must verify the member's current plan, network, benefit, prior-authorization instructions, delegated reviewer, portal, and response terms for the requested wound service.

  • Confirm plan and product identity, effective dates, provider and site participation, and whether the request is handled by the plan or a delegated entity.
  • Identify the applicable Traditional Medicare NCD or LCD context and the plan's current, publicly available criteria or instructions where permitted.
  • Keep skin-substitute, debridement, HBOT, site, product, and continuation questions separate; one approval label may not answer every downstream requirement.
  • Route adverse decisions, information requests, and continuity questions under the response's actual notice and appeal instructions rather than calculating rights from a generic template.

Human handoff: MA authorization and payer-relations staff verify current plan criteria and response scope; clinicians approve clinical support; network and revenue-cycle teams confirm operational and financial consequences. Qualified compliance or legal staff review disputed criteria, notices, continuity, or appeal rights.

Sources for this path: Centers for Medicare & Medicaid Services, Centers for Medicare & Medicaid Services Medicare Coverage Database

medicaid

Medicaid: state program and delivery system control the route

Medicaid coverage, documentation, prior authorization, product rules, provider enrollment, and appeal processes vary by state and may differ between fee-for-service and managed care. Federal managed-care authorities allow different state delivery structures, so a Medicare LCD or commercial policy is not a substitute for the current state manual, fee schedule, plan contract, portal, and notice.

  • Verify the state, eligibility and coverage date, fee-for-service or managed-care enrollment, responsible plan, provider status, benefit, and service location.
  • Use the state's and plan's current skin-substitute, wound-care, HBOT, product, prior-authorization, coding, and submission instructions when they apply.
  • Track state and plan response timeframes, notices, reconsideration, and appeal instructions from the actual source rather than assuming a universal federal operational deadline.
  • Preserve coordination-of-benefits and dual-eligibility questions for qualified staff without allowing the agent to sequence payers or determine liability.

Human handoff: State Medicaid and managed-care specialists verify the controlling program sources and portal; clinicians approve clinical evidence; enrollment, contracting, coding, and revenue-cycle teams resolve their domains. The state or plan makes coverage decisions, and qualified counsel handles contested rights.

Sources for this path: Medicaid.gov

commercial

Commercial coverage: the contracted plan policy is the evidence target

Commercial payers can publish detailed, product-specific skin and soft tissue substitute criteria and separate HBOT or wound policies, but one carrier's document is only an example of plan variation. The controlling evidence target is the current policy and contract for the actual payer product, service, product, provider, site, and date—not a generalized industry rule.

  • Confirm payer product, benefit, network, delegated reviewer, policy identifier, effective date, form, portal, and attachment channel.
  • Map the request to the actual product and indication language without treating another carrier's criteria or an expired policy as interchangeable.
  • Separate initial request, repeat application, change of product, HBOT course, information request, denial, and appeal workflows.
  • Treat authorization and benefits information as conditional administrative facts, not a guarantee of coverage, reimbursement, patient responsibility, or payment.

Human handoff: Contracting, patient-access, reimbursement, and authorization staff confirm the current contracted source and response. Clinicians control clinical support and care; coding and finance staff control claim and financial release; compliance or legal experts resolve disputed plan terms or notices.

Sources for this path: Aetna

workers comp auto liability

Workers’ compensation and auto liability: jurisdiction and causation stay human

Workers’ compensation, occupational injury, and auto-liability workflows can add jurisdiction, network, adjuster, utilization-review, causation, treatment-guideline, lien, and legal requirements that do not exist in a standard health-plan request. Texas's official preauthorization page is one jurisdictional example, not a national rule or an auto-liability template.

  • Confirm claim type, jurisdiction, carrier or administrator, network, employer or adjuster contact, provider status, and the controlling treatment and preauthorization source.
  • Keep health-plan coverage separate from compensability, causation, apportionment, lien, settlement, and third-party liability questions.
  • Route work-status, return-to-work, utilization-review, peer-review, and dispute instructions to authorized professionals under the actual jurisdiction.
  • Do not submit records, products, treatment, or claims based on a general health-plan authorization when liability-route approval is also required.

Human handoff: Workers’ compensation or liability specialists verify jurisdiction, network, adjuster, authorization, and billing requirements; clinicians address care and clinical causation only within their authority; compliance and legal counsel control disputes, liens, notices, and litigation strategy.

Sources for this path: Texas Department of Insurance, Division of Workers’ Compensation

Human authority, source provenance, and media restraint are product requirements

Wound evidence is clinically sensitive, visually identifying, financially consequential, and easy to strip from its original context. Governance therefore has to be visible in daily work: who can see an artifact, which source supports a statement, what changed, what an agent was allowed to do, who approved the next step, and how work continues when a model or connection is unavailable.

Role and decision-right separation

Configure distinct permissions for clinical meaning, record custody, authorization, product custody, coding, charge release, submission, appeal, privacy, and legal review. A single ready label cannot collapse those approvals, and no agent may approve its own output.

Field-level provenance and versioning

Retain source system, document, author, service or capture date, retrieval time, version, amendment, extraction method, reviewer, and approval state for material fields. Corrections create lineage; they do not erase the earlier packet or payer response.

Clinical and media non-interpretation boundary

Block image analysis, derived wound measurements, staging, diagnosis, treatment recommendation, product selection, and inferred response. Media workflows manage custody and disclosure only, while numeric conflicts and clinical meaning route to qualified clinicians.

Policy and response scope control

Require program, payer product, jurisdiction, service, provider, site, product, effective date, and source version before a rule can guide work. Record uncertainty and prevent an old authorization, neighboring LCD, or another plan's policy from being silently reused.

Minimum necessary and sensitive-data handling

Limit fields, artifacts, exports, recipients, and retention to the approved purpose. Apply encryption, access review, disclosure logging, and incident controls according to the buyer's legal and security analysis, including special attention to photographs and embedded metadata.

Quality monitoring and manual fallback

Test episode linkage, source accuracy, missing-evidence detection, escalation routing, packet versioning, response matching, and product reconciliation with approved synthetic or de-identified scenarios. Track false matches, missed holds, overrides, channel failures, and owner acknowledgement before expansion.

Human authority
Treating clinicians retain diagnosis, staging, measurement interpretation, care planning, debridement decisions, skin-substitute and product selection, HBOT decisions, clinical attestations, and urgency. Authorized payer, records, supply, coding, finance, compliance, privacy, security, contracting, and legal professionals retain their respective decisions. Agents organize evidence and route work; they do not replace those authorities.
Audit trail
The audit record captures source and version, agent action, confidence basis, open conflict, human reviewer, approval or override, packet and attachment version, submission event, acknowledgement, response linkage, change event, product-ledger reconciliation, downstream handoff, and closure reason. Buyers set retention, access, legal hold, export, and deletion rules.
Data boundary
Production workflows may process sensitive data only under the buyer's approved architecture, contracts, access model, and minimum-necessary configuration. Marketing analytics must never receive patient, photograph, wound, measurement, diagnosis, order, product, lot, payer, authorization, portal, claim, schedule, liability, or credential values. Post-publication GA4 measurement is limited to approved non-PHI route context, content cluster, and CTA label or location; Search Console review uses aggregate route-level clicks, impressions, CTR, position, and query themes.

Connect the evidence trail without pretending one system owns it all

Implementation starts with a source-and-authority map, not a promise of universal connectivity. Each connection has an approved purpose, minimum data set, responsible owner, failure state, and manual fallback. Source records remain authoritative; the workflow stores provenance and review state rather than silently rewriting clinical, inventory, authorization, or claim systems.

EHR, practice management, and scheduling

Information in scope

Approved encounter, referral, order, note, provider, site, schedule, and document metadata can establish episode context and retrieve signed records for review. Write-back is limited to buyer-approved tasks, links, and status references after field-level validation.

Boundary

The workflow does not alter signed clinical documentation, diagnose, stage, calculate measurements, select care, or release an appointment. Missing or stale data creates a visible task and preserves the source-system status.

Wound documentation, measurement, and media platforms

Information in scope

Approved measurements, method metadata, artifact identifiers, capture context, note links, and export status can support chronology and custody. The original platform remains the source for clinical values and images.

Boundary

No pixel analysis, image comparison, tissue recognition, derived measurement, or clinical trend is performed. Unsupported formats, lost provenance, or conflicting values are held for clinician, records, privacy, or security review.

Payer portals, clearinghouses, and document channels

Information in scope

Validated interfaces may exchange approved request fields, attachments, acknowledgements, information requests, and response documents when payer terms and buyer controls permit. Every transaction is linked to its exact packet version.

Boundary

A connector does not make a payer instruction authoritative, guarantee receipt, or turn a status label into coverage. Channel failure, ambiguous response, credential issue, or scope mismatch invokes the tested human fallback.

Inventory, procurement, and product custody

Information in scope

Buyer-approved product identifiers, purchase and receipt references, lot or batch records, available package context, use, discard, return, and wastage reason can enter the administrative ledger for reconciliation.

Boundary

The workflow is not inventory control and does not reserve, dispense, select, substitute, recall, or release a product. Supply and clinical owners approve all custody, availability, product-choice, and use decisions in their systems of record.

Charge capture, coding review, billing, and finance

Information in scope

Approved procedure, product-ledger, authorization-scope, charge, and claim-draft references can be compared before release. Exceptions remain assigned to coding, compliance, supply, clinician, or finance owners.

Boundary

The agent does not choose codes, modifiers, units, charges, write-offs, claim status, or patient responsibility. It cannot treat a matching ledger as proof that a service is billable, covered, medically necessary, or payable.

Identity, access, audit, and data governance

Information in scope

Role, organization, purpose, consent, retention, access, export, and audit controls can be applied to sensitive wound, photograph, payer, product, and claim evidence. Environment and user context remain buyer governed.

Boundary

Access is not inherited merely because a source can be connected. The buyer must approve minimum-necessary fields, roles, business-associate requirements, retention, deletion, monitoring, incident response, and use of any downstream vendor.

Model administrative capacity with inputs your team can replace

Use a transparent planning formula: monthly wound cases or administrative requests multiplied by administrative minutes saved per case, divided by sixty, multiplied by an illustrative loaded labor rate. Count only observed administrative handling time that the bounded pilot actually changes. Do not add assumed approvals, denials avoided, product revenue, clinical outcomes, reimbursement, or patient volume.

Monthly cases

180 cases

Illustrative volume for planning only. Replace it with the buyer's counted monthly cases in the selected wound, skin-substitute, HBOT, documentation-review, or reconciliation lane.

Administrative minutes saved per case

22 minutes

Illustrative time assumption for source retrieval, chronology assembly, question routing, packet versioning, and reconciliation. Validate it with observed before-and-after handling time during the pilot.

Illustrative loaded labor rate

38 dollars per hour

Illustrative wage-plus-burden input, not QuickIntell pricing or a market benchmark. Finance should replace it with the approved loaded rate for the roles whose administrative time is measured.

Formula

180 cases × 22 administrative minutes saved per case ÷ 60 × $38 illustrative loaded labor rate = $2,508 of modeled monthly administrative capacity.

Illustrative result

The illustrative formula yields $2,508 per month in administrative capacity. It is not revenue, cash savings, avoided denial value, reimbursement, or a forecast; use measured pilot inputs and show the result alongside quality, safety, and exception findings.

Illustrative planning model—not a customer result, guarantee, or quote.

Illustrative workflow examples

These examples explain process behavior. They are not customer stories, measured outcomes, clinical advice, or promises of coverage.

Illustrative example

A skin-substitute packet has serial records but no safe baseline chain

A wound center is preparing a skin-substitute request. The chart contains several signed notes, measurement entries, a photograph manifest, offloading references, and a vascular report, but the earliest measurement uses different anatomical wording and a later note repeats a treatment statement without an explicit current assessment. This is a fictional process example with no patient, payer, order, product, or claim data.

  1. The Wound Episode Identity Agent holds the earliest record apart until records staff and the clinician confirm that it belongs to the same wound episode.
  2. The Serial Wound Evidence Librarian displays the measurement methods and conflicting wording without calculating a trend or selecting a preferred value.
  3. The Conservative-Care Qualification Mapper separates documented offloading activity from the clinician-authored conclusion the payer question requests.
  4. The Clinical Media Custodian verifies artifact lineage and disclosure approval without examining the photograph, while the Packet Builder keeps all unresolved items visible.
  5. After qualified humans resolve identity, clinical meaning, current policy, and attachment scope, an authorized submitter approves the exact packet version.

Illustrative outcome: The illustrative outcome is a traceable packet or a justified hold, not an inferred qualification. Reviewers can see which source supports each approved answer, which human supplied clinical meaning, and why the earlier version did not proceed.

Illustrative example

An HBOT response no longer matches the scheduled course context

A hyperbaric program receives a payer response tied to an earlier request version. Before downstream release, the site context and scheduled course details change, while the wound timeline also receives an amended clinician note. This is a fictional administrative example and does not describe a real patient, payer, authorization, service, or clinical decision.

  1. The Response and Change-Control Coordinator links the response to the submitted version and identifies the fields that no longer agree with current operations.
  2. The prior response stays immutable; a new version reopens only the affected policy, clinical, site, scheduling, and financial checks.
  3. The Serial Wound Evidence Librarian records the amended note lineage without treating the amendment as proof that coverage criteria are met.
  4. Authorization staff verify the payer's current instructions, the clinician approves clinical support, and scheduling and revenue-cycle owners maintain a hold until their release conditions are satisfied.
  5. If the channel or response remains ambiguous, the team uses the tested manual payer-contact and documentation path rather than allowing the agent to infer scope.

Illustrative outcome: The illustrative outcome is a controlled correction path with the old request, response, amendment, new evidence, human approvals, and downstream acknowledgements connected. The example does not promise authorization, treatment timing, coverage, or payment.

Start with one evidence-heavy lane and earn the right to expand

Adoption should begin where the current process, authorities, source systems, and exception burden can be observed. A bounded rollout proves linkage accuracy, human control, media restraint, payer-source freshness, product reconciliation, and manual fallback before more sites, services, products, payers, or automated actions are added.

  1. 1. Map the current lane and decision rights

    • Choose one wound workflow, location, payer route, and accountable executive owner
    • Document the before-state from intake through packet, response, procedure, product, charge, and claim handoff
    • Name every clinical, records, authorization, supply, coding, privacy, finance, compliance, and legal decision owner
    • Inventory authoritative sources, current policies, connections, media flows, and manual fallbacks

    Exit criteria: The team approves a non-PHI process map, source-and-authority matrix, explicit non-goals, initial integration boundary, escalation service levels, and measurable acceptance criteria. Clinical interpretation and autonomous coverage or claim decisions are out of scope.

  2. 2. Configure and test with approved scenarios

    • Configure episode identity, chronology, payer route, evidence categories, approval gates, and version rules
    • Validate field-level provenance and minimum-necessary access for notes, measurements, media, payer, and product records
    • Exercise identity conflicts, missing baselines, stale policies, media holds, quantity mismatches, ambiguous responses, and connection failures
    • Train each role on approve, correct, hold, override, escalate, and manual-fallback actions

    Exit criteria: Approved synthetic or de-identified scenarios demonstrate that required holds cannot be bypassed, sources remain traceable, photographs are never interpreted, every high-risk action requires the right human, and outages return work to a usable manual queue.

  3. 3. Run a supervised pilot

    • Operate the new workflow beside the current control for the selected lane
    • Measure administrative touch time with the disclosed cases-times-minutes-times-loaded-rate model
    • Review linkage errors, missed evidence, false holds, overrides, response mismatches, product exceptions, and owner acknowledgement
    • Collect role-specific feedback without placing PHI or operational values in marketing analytics

    Exit criteria: The steering group accepts documented evidence quality, human-review performance, privacy and security findings, manual fallback, training readiness, and measured administrative inputs. Open safety, provenance, policy, or product-control defects block expansion.

  4. 4. Expand under change control

    • Add one payer, location, service, product family, or integration at a time
    • Revalidate current policies, contracts, payer instructions, and source ownership before each addition
    • Monitor drift, access, model behavior, integration failures, policy staleness, and unresolved exceptions
    • Review governance, retention, business continuity, and vendor-risk controls on the buyer's schedule

    Exit criteria: Each expansion has a named owner, approved source set, tested scenarios, trained users, performance baseline, rollback path, and governance sign-off. A successful pilot in one lane is not assumed to validate another payer, service, product, or jurisdiction.

Authoritative sources used for coverage context

Sources support the cited coverage and administrative context. Teams must confirm the current policy, contract, jurisdiction, and effective date for each real case.

  1. Final Local Coverage Determinations for Certain Skin Substitutes Withdrawn

    Centers for Medicare & Medicaid Services · government · reviewed

    CMS states that its A/B MACs withdrew the skin-substitute LCDs scheduled to take effect on January 1, 2026. This supports the warning that teams must verify current active local sources and must not treat a withdrawn future policy as a nationwide rule.

  2. LCD L35041: Application of Bioengineered Skin Substitutes to Lower Extremity Chronic Non-Healing Wounds

    Centers for Medicare & Medicaid Services Medicare Coverage Database · official payer policy · reviewed

    This active local policy is an authoritative example of jurisdiction-specific skin-substitute coverage and documentation language, including serial measurement and conservative-care context. It is cited as a local example, not as a universal Medicare or commercial rule.

  3. WISeR (Wasteful and Inappropriate Service Reduction) Model

    Centers for Medicare & Medicaid Services Innovation Center · government · reviewed

    CMS describes the active Original Medicare model, its selected states and services, the provider choice between prior authorization and prepayment review for included services, its human clinical-review boundary, and that the model does not change coverage or apply to Medicare Advantage.

  4. National Coverage Determination 20.29: Hyperbaric Oxygen Therapy

    Centers for Medicare & Medicaid Services Medicare Coverage Database · official payer policy · reviewed

    The NCD provides Medicare's national hyperbaric oxygen coverage language, including listed conditions and the standard-wound-therapy and continuing-evaluation provisions for certain diabetic lower-extremity wounds. Qualified staff must apply it to real facts and current local instructions.

  5. Contract Year 2026 Policy and Technical Changes to the Medicare Advantage Program

    Centers for Medicare & Medicaid Services · government · reviewed

    CMS summarizes existing guardrails for Medicare Advantage use of Traditional Medicare coverage criteria and internal coverage criteria. It supports a separate MA path while current regulations, plan documents, delegated arrangements, and responses still require qualified review.

  6. Managed Care Authorities

    Medicaid.gov · government · reviewed

    Medicaid.gov explains that states operate managed care through different state-plan and waiver authorities and that federal regulations include appeal and grievance protections. This supports state- and plan-specific routing rather than reuse of Medicare or commercial rules.

  7. Clinical Policy Bulletin 0244: Skin and Soft Tissue Substitutes

    Aetna · official payer policy · reviewed

    This current official carrier policy shows that a commercial payer may publish detailed, product- and indication-specific wound evidence and continuation criteria. It is used only to demonstrate payer variation; it does not govern another plan or guarantee coverage.

  8. Preauthorization and Treatment Guidelines

    Texas Department of Insurance, Division of Workers’ Compensation · government · reviewed

    Texas identifies different network and non-network preauthorization sources and instructs providers to check network contracts and applicable rules. It supports jurisdiction-specific liability routing and is not presented as a national workers’ compensation or auto-liability rule.

  9. Use of Online Tracking Technologies by HIPAA Covered Entities and Business Associates

    U.S. Department of Health and Human Services Office for Civil Rights · government · reviewed

    OCR explains that HIPAA-regulated entities must evaluate tracking technologies and may not impermissibly disclose PHI. This supports the page's non-PHI analytics contract, while each buyer remains responsible for fact-specific legal, privacy, security, and vendor review.

  10. Tissue and Tissue Product Questions and Answers

    U.S. Food and Drug Administration · government · reviewed

    FDA explains its risk-based regulation of human cells, tissues, and cellular and tissue-based products and notes that not every HCT/P follows the same premarket pathway. This supports product-specific human review and avoids treating all skin substitutes as one regulatory class.

Wound Care and Hyperbaric Medicine workflow FAQs

Open a question to review the answer. The disclosure controls use native browser behavior and remain available by keyboard without page JavaScript.

Why is this different from the wound module already in our EHR?

The EHR remains a source of truth for signed clinical documentation. This workflow connects evidence across EHR notes, wound measurement or media systems, payer instructions, product custody, authorization responses, charge capture, and claim review. It adds source provenance, question-specific gaps, human approvals, and change lineage without replacing the EHR or editing clinical meaning.

Does the software measure wounds or interpret serial photographs?

No. It can index a source-system value and manage photograph identity, custody, linkage, rendition, consent, redaction, and disclosure state. It does not inspect pixels, calculate dimensions or area change, recognize tissue, assign stage, compare appearance, or infer healing. A qualified clinician reviews every image and clinical measurement.

Can an agent decide that conservative care failed or that a patient qualifies for a skin substitute?

No. The agent maps dated records to the current human-selected question and separates documented activity from explicit clinician-authored meaning. It cannot determine adherence, response, failure, contraindication, medical necessity, qualification, or coverage. The treating clinician supplies clinical conclusions, and the payer retains coverage authority.

Does it recommend a skin-substitute product or application plan?

No. Product selection, clinical use, application frequency, substitutions, and treatment planning remain with qualified clinicians and approved organizational processes. The product ledger can reconcile the human-approved request with procurement, lot or batch, use, discard, wastage reason, procedure note, charge, and claim draft, then route mismatches to accountable humans.

How does the workflow handle the withdrawn Medicare skin-substitute LCDs and WISeR?

It does not encode one permanent Medicare checklist. Staff record the current MAC, jurisdiction, active MCD source, service, date, and whether WISeR applies. CMS withdrew certain LCDs scheduled for January 2026, while pre-existing local policies may remain active; WISeR is limited to Original Medicare, selected states, and selected services. Qualified staff verify every case.

Can the same evidence record support a hyperbaric oxygen request?

The governed wound chronology may be reused after source and scope review, but HBOT receives its own request, current coverage-source map, clinical approvals, packet, response, and change controls. The agent does not decide whether NCD 20.29 or another current source is satisfied, recommend HBOT, classify a wound, or determine continuation.

How are debridement depth, procedure count, lot tracking, and wastage handled?

The workflow preserves the signed clinician-authored procedure facts and compares them with approved inventory, lot or batch, quantity, discard, wastage-reason, charge, and claim-draft records. Conflicts create named pre-bill work. It never infers debridement depth, changes a note, invents wastage, or selects codes, modifiers, units, or charges.

What happens when the payer, site, product, provider, or planned service changes?

The prior packet and response remain immutable. A new version identifies which assumptions changed and reopens only the affected clinical, coverage, authorization, product, scheduling, coding, or financial checkpoints. No prior ready or approved label silently carries forward, and authorized humans decide whether correction, resubmission, new authorization, hold, or another path is required.

Do you need direct access to every payer portal and wound system?

No. Start with the smallest reliable connection set for one lane. Validated APIs, files, document feeds, or buyer-approved manual steps can all participate if provenance, access, acknowledgement, failure behavior, and ownership are explicit. Missing connectivity does not justify screen scraping, credential sharing, unsupported automation, or a false submitted status.

How do you keep photographs and wound details out of marketing analytics?

The page measurement contract permits approved non-PHI route context, content cluster, and CTA label or location only. It excludes patient, photograph, wound, measurement, diagnosis, order, product, lot, payer, authorization, portal, claim, schedule, liability, and credential values. Buyers must separately review production data flows, vendors, access, contracts, and HIPAA obligations.

Can the agent write and submit an appeal after a denial?

It can organize the authenticated denial, submitted packet, current instructions, source-linked approved evidence, missing decisions, and accountable deadlines for human review. It cannot choose a legal or clinical strategy, calculate a disputed deadline from incomplete facts, author unsupported clinical meaning, sign, attest, or file an appeal without the buyer's required qualified approvals.

What does the implementation offer include, and is the software free?

The approved offer is $0 implementation fee. $0 customization charges. Software subscription and usage charges are separate, so the product itself is not free. The commercial agreement should define scope, dependencies, transaction or third-party charges, connections, data migration, support, recurring fees, and usage assumptions before the buyer begins a pilot.

Review connected product and specialty workflow pages without adding this page family to global navigation.

Bring one difficult wound-evidence lane to a working session

Choose one location, payer path, wound workflow, and non-PHI process map. We will identify the before-state, evidence sources, policy boundaries, media controls, product and claim handoffs, human decision rights, integration limits, manual fallback, adoption gates, and transparent labor-capacity formula. Do not send patient, photograph, wound, order, product, authorization, claim, portal, or credential data through the marketing form.