DME & HME automation: frequently asked questions

What is DME automation?

DME automation is software that runs the repetitive back-office workflows a DME/HME platform does not orchestrate on its own — intake, prior authorization follow-up, resupply outreach, proof-of-delivery chasing, and pre-claim documentation checks. It is not a replacement for your billing platform. It is a layer that sits on top of Brightree, WellSky, NikoHealth, or a homegrown system and does the clicking, checking, and following up that staff currently do by hand.

The practical test: if a task is high-volume, rule-based, and generates no revenue by itself, it is an automation candidate. If it requires clinical judgment or a payer negotiation, it is not.

Which DME workflows are worth automating first?

The highest-ROI workflows are the ones with the worst manual-to-revenue ratio. In order:

• Fax and referral intake — OCR plus field extraction, sender and patient matching, routing to the right queue.
• Prior authorization tracking — submission status polling, expiry warnings, and re-auth triggers before the auth lapses.
• Resupply outreach — payer-specific eligibility cadences with automated patient contact and response capture.
• Proof of delivery — chasing missing PODs before they become unbillable claims.
• Pre-claim documentation validation — catching missing CMNs, signatures, and chart notes before submission rather than after denial.

These are precisely the gaps all-in-one platforms leave open, because platforms are built to record the transaction, not to chase the paperwork around it.

Does AI replace human review in DME operations?

No, and any vendor who says otherwise should worry you. In a compliant DME workflow the automation handles extraction, matching, routing, and follow-up; a human approves anything that affects a claim, a clinical record, or a patient communication of consequence.

The standard pattern is confidence-thresholded: documents the model reads cleanly flow straight through, and anything ambiguous is escalated to a staff queue with the extracted fields pre-filled. Staff stop doing data entry and start doing exception handling — which is both faster and more defensible in an audit.

Can automation integrate with Brightree?

Yes. Brightree automation is typically built through a combination of Brightree's documented APIs, HL7/FHIR interfaces where the data is clinical, secure read-only database access for reporting, and UI-level automation (RPA) for the specific screens that expose no API.

Brightree remains the system of record throughout. The automation layer reads referrals and order status out, and writes authorization status, documentation flags, and follow-up outcomes back in. Nothing migrates and no one relearns a new interface — the staff-facing change is that queues arrive pre-populated instead of empty.

The same approach applies to WellSky CareTend, Bonafide, and NikoHealth. The integration method changes; the architecture does not.

Do I have to replace my DME platform to get automation?

No — and for most suppliers, replacing is the more expensive mistake. A platform migration is a 6–18 month project with data conversion risk, retraining cost, and a revenue dip during cutover. An automation layer on top of your current platform is typically a 4–8 week build with no migration at all.

Replacement is the right call in one situation: when the platform itself genuinely cannot support your business — it is unsupported, it lacks a payer or product line you need, or the vendor is exiting the market. Short of that, automating on top preserves your historical data and your staff's muscle memory.

Can we keep our existing systems and still automate?

Yes. That is the entire premise of an automation layer. Your DME platform, your fax server, your document store, and your accounting system all stay exactly where they are. The automation connects them and runs the workflows that currently require a human to move information between them.

This matters most for suppliers running a mix — a modern billing platform, a legacy fax line, and a spreadsheet somebody maintains for resupply. Automation does not require you to consolidate first. It is usually the cheapest way to make a mixed stack behave like one system.

Do you help with data migration from our current system?

Yes, when migration is genuinely the right answer — but we will tell you plainly when it is not. If you are moving platforms for a real reason, the work covers extraction, field mapping, cleanup of duplicate and malformed patient records, and reconciliation testing before cutover.

If you are considering migration mainly because your current platform “does not automate,” that is usually solvable without moving, and we will say so.

Can DME providers automate fax order intake?

Yes, and it is usually the first workflow worth automating because faxed referrals are the highest-volume, lowest-judgment work in a DME operation. Automated intake receives the fax, cleans and de-skews the image, classifies the document type, extracts patient and referral fields via OCR, matches the patient against your platform, and routes it to the correct queue.

What staff keep: reviewing exceptions and anything the model flagged as low-confidence. What staff stop doing: retyping demographics off a fax.

The practical gate is fax quality. Clean digital faxes extract reliably; a stack of third-generation photocopies will produce more exceptions and needs realistic expectations set before the build.

How does prior authorization automation actually work?

Prior auth automation does not submit and forget — it tracks. The system watches every open authorization, polls or checks status on a schedule, flags auths approaching expiry before they lapse, triggers re-authorization workflows, and escalates anything stalled past a threshold you set.

The revenue effect comes from eliminating the two expensive failure modes: delivering against an expired auth, and letting an approved auth sit unworked until it expires. Both are silent — neither shows up until the denial does.

You can estimate the time impact for your own volume with the prior authorization time savings calculator.

How does CPAP resupply automation work?

Resupply automation runs eligibility on payer-specific cadences and contacts eligible patients automatically by text, email, or voice, then captures the response and creates the order. The core problem it solves is that manual resupply outreach scales with headcount — so most suppliers only ever reach a fraction of their eligible panel.

Because eligibility rules differ by payer and product, the cadence logic has to be configured per payer rather than run as a single global schedule. That configuration is most of the build effort.

How much does resupply automation actually recover?

On SynergyIQ builds, resupply and documentation automation has recovered roughly $210,000–$280,000 per 2,000-patient panel, driven mostly by contacting eligible patients who were never reached manually and by preventing denials from missing documentation.

Treat that as a range, not a promise. The figure depends heavily on your current outreach coverage, payer mix, and denial rate. A supplier already running disciplined manual outreach will see a smaller lift than one reaching 40% of its eligible panel. You can model your own numbers with the DME revenue leak calculator.

How can automation reduce order-processing delays?

Delay in DME is almost never caused by the processing step itself — it is caused by waiting. A referral sits in a fax queue overnight. An auth status goes unchecked for three days. A missing chart note is discovered at billing instead of at intake.

Automation compresses the waiting, not the working. Intake is classified within minutes of arrival instead of at the next queue sweep, auth status is checked on a schedule rather than when someone remembers, and documentation gaps surface at intake when they are cheap to fix rather than at claim submission when they are not.

How should a DME or HME provider begin an automation project?

Start with one workflow that is high-volume, rule-based, and measurable — not with a platform-wide transformation. The sequence that works:

• Measure the bleed. Pick the workflow costing the most staff hours or the most denied claims. Get a real number before you build anything.
• Automate one workflow end to end. Usually fax intake or prior auth tracking. Ship it, measure it, confirm the number moved.
• Expand to adjacent workflows once the first one is demonstrably working and staff trust it.
• Only then consider whether the underlying platform still needs to change — often it no longer does.

The common failure is starting with a multi-workflow build before anyone has proven the first one saves time. That produces a long project with no early evidence, which is how automation initiatives lose internal support.

What does DME automation cost?

SynergyIQ prices automation as fixed-scope projects rather than open-ended hourly work. A single workflow typically runs $2,000–$8,000. A multi-workflow operational build typically runs $10,000–$40,000.

What moves the number within those ranges:

• Integration method. A documented API is cheapest. HL7/FHIR is moderate. RPA against a UI with no API is the most expensive to build and maintain.
• Number of payers. Payer-specific rules multiply the configuration work — this is usually the single largest driver.
• Document quality. Clean digital faxes are straightforward; poor scans need more exception handling.
• Compliance scope. PHI-handling workflows require a BAA, audit logging, and access controls that add build time.

How long does a DME automation build take?

A single-workflow build typically takes 4 to 8 weeks from kickoff to production, with payback usually landing in the 60 to 90 day range. Multi-workflow builds run longer but are sequenced so the first workflow goes live early rather than everything landing at the end.

The schedule risk is rarely engineering. It is access — getting API credentials, a test environment, and payer documentation. Projects that stall usually stall there, which is why we front-load access requests in week one.

Do you sign a Business Associate Agreement?

Yes. Any workflow touching protected health information runs under a signed BAA before work begins — that is a HIPAA requirement for a business associate, not an optional courtesy.

Practically it also means PHI stays inside systems covered by that agreement, access is least-privilege and logged, and we do not move patient data into tools that are not covered. See the healthcare IT and HIPAA FAQ for how this works across the rest of the stack.