Best Referral Processing System: 3 Steps to Choose in 2027
Short answer: pick the system that closes the loop
For most small healthcare businesses, the right referral processing system is the one that tracks each referral to a booked and completed visit, not the one with the longest feature list. A referral processing system is software that takes in a referral, pulls out its key details, routes it to the right person, and keeps it open until the patient is seen or the referral is formally closed. Shortlist on closed-loop tracking and EHR fit first, then ask for a written quote, because none of the six vendors below shows a dollar price on the pages we opened.
Key Takeaways
A referral processing system is worth its cost only if it tracks referrals to a completed visit. Intake alone leaves the stall in place.
None of the six vendors shows a dollar price on the pages we opened. Phreesia publishes a free access window and a standard start date, and the rest are quote-based or show no price at all.
Luma Health and Tennr describe fax-driven intake in the most detail. Mosai is built for home health and hospice. Phreesia fits best if you already run its intake.
A 2018 primary care study found that 34.8% of referral scheduling attempts ended in a documented completed appointment.
Zapier, Make and n8n can run this workflow, but you own duplicate control, escalation, access rules and upkeep.
Run every shortlisted vendor through the same demo: one real referral from intake to a booked visit, with the human review step visible.
How we evaluated these six referral tools
This is an informed buyer's guide built from public information. We did not run trials or test any product, and we did not contact vendors. We reviewed vendor pages, a vendor-authored comparison, a trade-press award list, and peer-reviewed research. The weights below are our analysis, not vendor or market data.
| Criterion | Weight | Why it carries that weight |
|---|---|---|
| Referral-to-visit tracking (scheduling and loop closure) | 25% | Referrals stall at scheduling, so a system that stops at intake leaves the main gap open |
| Intake from fax, email and portals | 20% | Most referrals still arrive as documents that must be sorted and labeled |
| EHR or practice-management connection | 20% | Without a named connection, staff re-key data and records drift |
| Human review, audit and access control | 15% | Clinical and privacy decisions need a visible reviewer and a log |
| Pricing transparency and total cost | 10% | Quote-based pricing is common, but the quote must cover implementation |
| Fit for a small outpatient business | 10% | Home health, hospice or enterprise-only products add scope you may not need |
The weights sum to 100%. A vendor with a strong intake demo but no visible scheduling step scores lower under this weighting, and that is deliberate.
Who this is for
This guide is for an owner or operations lead at a small healthcare business who receives referrals by fax, portal, phone or email and is close to choosing a system. The fit is strongest if you have an EHR or practice-management system with an export or API connection, and one named person who owns the referral queue.
Red flags: you run home health or hospice and need that focus, since Mosai's site describes home health and hospice EHR integrations; you have no EHR or practice-management system to connect to; or you need a published price before any sales conversation, which most vendors here do not offer.
What the public pages actually show
The matrix below records only what we could read on each vendor's pages. "Not stated" means the page did not say, not that the feature is missing.
| Product | Referral intake | Scheduling and loop closure | EHR connections named | Human review described | Public price on page |
|---|---|---|---|---|---|
| Phreesia | Referral handling named on homepage | Not explained | Not named | Not described | Quote-based |
| Luma Health | Yes, through Fax Transform | Yes: creates the EHR referral, coordinates scheduling, closes the loop | 5 named, plus 4 partners | Yes: confidence thresholds set per document type | Not shown |
| Assort Health | Referral requests handled (vendor copy) | Scheduling coordination claimed | Athena integration claimed | Not described | Not shown |
| Tennr | Universal Inbox from eFax, email, e-portal and e-prescribe | Scheduling not named | General EHR claim only | Not described | Not shown |
| Mosai | AI data extraction and prioritization | Not described | Home health and hospice EHRs | Not described | Not shown |
| ReferralMD | Positioned as referral management and network (page title only) | Not verified | Not verified | Not verified | Pricing page showed only its title |
Two rows need caution. Luma's referral coordination sits on a fax product page, so ask whether it can be bought as a standalone referral module. Forcura's domain now redirects to Mosai's site, and Mosai's site links to "Forcura + Medalogix," so confirm which product and contract entity you would sign.
Where referrals stall, and what the evidence says
Referral loss is measurable, but the numbers depend heavily on the setting. A 2018 study of a large primary care network analyzed fiscal 2016 data on 103,737 referral scheduling attempts, and Scheduling completion: 34.8% of attempts according to PubMed Central (2018). Many of the unsuccessful attempts lacked an appointment date, which is the step a referral processing system has to own.
A second 2018 study, of an in-house EHR at Partners HealthCare with an enhanced referral manager module, reported that 76.8% of 1,341 initiated referrals were completed, according to PubMed Central (2018). Those two rates come from different settings and definitions, so neither is a benchmark for a small practice. The useful lesson is that a system has to measure completion, not just intake.
A proposed US Tech Automations intake workflow, configurable for your own inbox, would start when a referral fax or emailed packet lands in a monitored queue. It would extract patient name, date of birth, referring provider, requested service and urgency into a draft referral record, and any field below a confidence threshold you set would go to a staff review queue before anything is written to the EHR. The output is a referral record with a status field, a dated outreach task for the scheduler, and an alert if no booking appears within a window you choose. Prerequisites are read access to the fax or referral inbox, an API or scheduled export from the EHR or practice-management system, and a business associate agreement with every vendor that touches patient data. The human review points are a staff member approving low-confidence fields and any urgent referral before outreach begins.
Six options, profiled
Each profile separates what the vendor's pages show from our read on fit. Where a page shows no price, we say so rather than estimate one.
Phreesia
Best fit: practices that already use Phreesia's intake at the front desk and want referral handling added to the same workflow. Phreesia's homepage pairs referral handling with fewer no-shows and easier booking, but it does not expand the referral wording into workflow steps. Evidence is the pricing page, according to Phreesia.
Limitations: the pages we opened do not explain referral mechanics, name EHR connections, or describe human review. Ask for a demo of referral handling specifically, not the intake product in general.
Implementation: the banner offers full access to the Intake platform at no cost through spring 2027, but the FAQ says products are included through May 31, 2027. Get the end date and the standard-pricing start in writing.
Luma Health
Best fit: outpatient practices that receive referrals by fax and run an EHR on Luma's named list. Luma's referral-management page describes Fax Transform, which creates the referral in the EHR, coordinates scheduling, and closes the referral and communication loop, according to Luma Health.
Limitations: the page lists EHR integrations with athenahealth, eClinicalWorks, Epic, Oracle Health and MEDITECH, plus four partners (NextGen, Nextech, Greenway Health and PCC). Those are listed for the product, not confirmed for your system, and the page does not say whether each applies to referral coordination.
Implementation: staff review is driven by confidence thresholds you set per document type, and referrals can be routed to pools or teams. No price is shown on the page.
Assort Health
Best fit: groups that want referral handling tied directly to booking, especially if Athena is the system of record. Assort's homepage positions its referral product around turning referrals into booked appointments and claims a deep Athena integration.
Limitations: the vendor's own June 2026 comparison is marketing content. It lists six platforms with Assort first, according to Assort Health (2026). The pages we opened do not describe discrete workflow steps or human review.
Implementation: no price is shown. Get the Athena scope and the review steps in writing before comparing quotes.
Tennr
Best fit: high-volume intake where most of the work is sorting faxes and documents across many patients. Tennr's referral-processing page describes a "Universal Inbox" fed by eFax, email, e-portal or e-prescribe, with sorting, labeling and prioritization. It also makes a claim we report as the vendor's own: Vendor claim: 30% of patients drop off according to Tennr.
Limitations: the described steps lean toward eligibility and benefits review. The page describes no human review step and does not name a scheduling system. Its EHR claim is general, so ask which EHRs are live today rather than planned.
Implementation: no price is shown, and the homepage's only call to action is a demo request.
Mosai
Best fit: home health and hospice providers. The 2026 Best in KLAS list names "Mosai Forcura" in patient referral management and analytics, as reported by HIT Consultant (2026).
Limitations: Mosai's site describes AI-powered referral intake and prioritization, and integrations with home health and hospice EHRs, naming Homecare Homebase, MatrixCare and Axxess. It does not describe scheduling or outpatient specialty use, so do not assume either.
Implementation: no price is shown. Confirm which product and entity your contract would name, given the Forcura redirect.
ReferralMD
Best fit: specialty practices that want a referral network as well as referral management. The site title positions ReferralMD as a referral management and healthcare referral network.
Limitations: the pricing page we opened returned only its title, so we could not confirm features, price or trial terms from the vendor's pages, according to ReferralMD. Third-party directories disagree about whether a starting price is published, so ask the vendor directly rather than relying on a listing.
Implementation: treat it as quote-based until the vendor confirms plan, trial and integration terms in writing.
Pricing and total cost
Pricing checked November 14, 2026. None of the six vendors shows a dollar price on the pages we opened, so the table records quote-based status and whatever access terms do appear.
| Product | Public price shown | Access terms on page | What to add to the quote |
|---|---|---|---|
| Phreesia | Quote-based | Intake access free through spring 2027 per banner; FAQ says May 31, 2027 | Implementation scope and the date standard pricing starts |
| Luma Health | Quote-based (none shown) | Not stated on referral page | Fax Transform scope, EHR connector fees, reviewer seats |
| Assort Health | Quote-based (none shown) | Not stated | Athena integration scope, referral workflow fees |
| Tennr | Quote-based (none shown) | Demo request is the only call to action | Connector scope for each fax or inbox source, support tier |
| Mosai | Quote-based (none shown) | Not stated | Contracting entity, product name, EHR connectors for your setting |
| ReferralMD | Quote-based (not confirmed) | Pricing page showed only its title | Written confirmation of plan, trial and integration terms |
The clearest date on any vendor page is the Phreesia standard-pricing start: Standard pricing start: May 1, 2027 according to Phreesia. Total cost of ownership is the quote plus staff time for review and exceptions, any connector fees, and the work of keeping an integration current when an EHR update changes a field. None of the six vendors gives a figure for those items on the pages we opened, so ask for them in writing.
Build it yourself or buy: Zapier, Make, or n8n
The alternative most buyers overlook is stitching this together in a workflow tool such as Zapier, Make or n8n, or building it in-house. Those tools can support run histories, retries, error branches and audit evidence when they are configured. The buyer still designs and owns the parts that cause referral failures: idempotency (a retried fax should not create two referral records), escalation when a referral sits untouched, access controls over patient data, and maintenance when a fax format or EHR field changes.
A proposed design from US Tech Automations would configure those pieces differently: a duplicate check before any record is created, an error queue with a named owner, escalation timers on referrals without a booking, role-based queues so schedulers see only their work, and a step-level log showing who approved each field. The prerequisites are the same as above, plus a named owner for the error queue and a clinician who sets the urgency rules. Human review stays in place, with clinical triage handled by a clinician and a staff member signing off on a regular review of stalled referrals. This describes a configurable design, not a deployed system.
Choose by situation
| Your situation | Shortlist first | Confirm before you buy |
|---|---|---|
| Home health or hospice referrals | Mosai | Contracting entity and scheduling scope |
| Fax-heavy outpatient intake with a named EHR | Luma Health | Whether Fax Transform is sold as a referral module |
| Front desk already on Phreesia intake | Phreesia | Referral mechanics and the end date of free access |
| High volume of mixed documents from many sources | Tennr | Live EHR list and human review steps |
| Athena-based scheduling | Assort Health | Athena scope and workflow steps in writing |
| Specialty referral network | ReferralMD | Features, trial terms and a written quote |
When NOT to use US Tech Automations: a packaged vendor is the simpler choice when referral volume is low enough that a shared inbox and a checklist handle it, when your EHR already includes a referral module you have not switched on, or when you want a vendor-hosted product with certified connectors for your EHR and no internal ownership of workflows. In those cases, start with the EHR's built-in referral queue or the packaged vendor whose scope matches your situation in the table above.
Worked example: 120 referrals a month
Take an illustrative practice that receives 120 referrals a month. At the 34.8% scheduling-completion rate from the 2018 primary care study cited above, about 42 referrals (120 × 0.348 = 41.76) end in a documented visit, which leaves about 78 without one. If a referral processing system lifted completion to a hypothetical 60%, 72 referrals would complete (120 × 0.60), a gain of about 30 booked visits a month. If the referral queue lives in HubSpot, its webhook documentation lists 8 subscription actions, including ASSOCIATION_ADDED, which a workflow could subscribe to when a referral record is linked to a patient contact, according to HubSpot Developers. The 60% is a target, not a benchmark, and the 34.8% comes from a different network, so use this math to set your own baseline rather than to forecast results.
Three steps to choose one
Map your referral path. List every way a referral arrives (fax, portal, phone, email), where it is sorted, who schedules it, and where the EHR record is created. Define "closed" as a booked, completed visit or a documented decision not to proceed.
Score the shortlist on the weighted criteria. Give each vendor the same sample referral with identifiers removed, and ask it to show that referral from intake to booked visit, including the human review step and the audit log.
Get the quote in writing. Ask for implementation scope, every connector and fee, the end date of any free access period, a business associate agreement, and the exact product and entity named on the contract. Then pilot one referral queue before you widen it.
Frequently asked questions
Is a referral processing system the same as an EHR referral module?
Not always. An EHR referral module usually works inside one record system, while a referral processing system can cover intake from fax, portal and email as well as scheduling. Check whether your EHR's module handles documents from outside sources before you buy a separate product.
Do most of these vendors publish prices?
No. None of the six showed a dollar price on the pages we opened. Phreesia publishes a free access window and a standard start date, and the rest are quote-based or show no price.
Why do referrals stall after they are sent?
The stall is usually at scheduling. In the 2018 primary care study, many scheduling attempts lacked an appointment date, which points to a step nobody owns rather than a clinical failure.
Can Zapier, Make or n8n replace a packaged system?
They can, if you accept that you design and maintain duplicate control, escalation, access rules and audit. A practice with an in-house developer can make that work; one without should buy a packaged product or pair one with its EHR module.
What should a demo prove?
A demo should prove one real referral moving from intake to a booked, completed visit, with the human review step and audit trail visible. A product tour without that path is not evidence of fit.
Where to go from here
Pick the system that tracks a referral to a completed visit, get scope and fees in writing, and test it on one of your own referral queues before you commit. To see how a configurable intake and tracking design could be set up for your inbox and EHR, see how US Tech Automations configures this.
About the Author

Helping businesses leverage automation for operational efficiency.