Healthcare revenue cycle
The Triune Health Partners RPA case study does not exist. Here is what the search is really about.
People type this looking for a named UiPath, Automation Anywhere, or Blue Prism deployment at a company called Triune Health Partners, with hours-saved numbers. There is no public case study by that name. The real company is TRIUNE Health Group, a workers’ compensation case management firm in Lombard, Illinois, and its revenue cycle is a genuinely harder automation problem than the hospital billing most of these stories describe. Verified June 2026.
Direct answer (verified June 22, 2026)
No public UiPath, Automation Anywhere, or Blue Prism case study names “Triune Health Partners.” Searching the three vendors’ case study libraries and the open web in June 2026 returns nothing under that name. The closest real entity is TRIUNE Health Group, a family-owned health care management company in Lombard, Illinois, whose business is workers’ compensation case management, not hospital billing or a published bot deployment.
If you are scoping automation for a case management or bill-review operation like that, the useful question is not “what did the case study say.” It is “which steps have no API, and how do you automate those without a brittle bot.” That is the rest of this page.
Who TRIUNE Health Group actually is
Pulled from triunehealthgroup.com: a Lombard, Illinois firm, family-owned for more than 35 years, working the workers’ compensation arena. None of these lines of business is hospital revenue cycle. They are clinical and vocational services, each of which generates its own billing and documentation trail.
Medical case management
Nurses coordinate care for injured workers, field and telephonic, from injury to return to work.
Nurse triage and catastrophic case management
First-report-of-injury triage and long-horizon management of severe, high-cost claims.
Utilization review
Clinical review of whether requested treatment is medically necessary under the claim and jurisdiction.
Vocational rehab and life care planning
Return-to-work plans, vocational assessments, and long-term cost-of-care projections.
Why a workers’ comp revenue cycle is the harder one to automate
Most healthcare RPA case studies describe hospital revenue cycle management: one payer per claim, an 837 going out to a clearinghouse, an 835 coming back, denials landing in a billing application. That world has standards. EDI formats exist precisely so that claims can move between systems without a human retyping them.
Workers’ compensation breaks that shape. A single bill for an injured worker can route through a third-party administrator, the employer, a state fund, and a separate bill-review vendor, each with its own portal login and its own jurisdictional fee schedule. There is no one clearinghouse that covers all of it. The case management platform, the bill-review tool, and the state filing portal usually expose no API to an outside automation. So the genuinely repetitive work, reading a value off one screen and keying it into another, is exactly the work with no programmatic shortcut.
Toggle the two below. The left is the textbook hospital cycle the existing case studies assume. The right is the multi-portal, no-clearinghouse reality of the workers’ comp world a firm like TRIUNE operates in.
Hospital RCM versus workers' comp revenue cycle
One payer relationship per claim. Claims and remits move as 837 and 835 EDI files through a clearinghouse. Denials drop into a single billing application. There is at least a standardized pipe between systems, even when the desktop client is still worked by hand.
- EDI standards (837 / 835) move data between systems
- A clearinghouse covers most payers
- One billing application as the system of record
The part that decides everything: how you reach a no-API screen
Generic RPA reaches those portals the brittle way: by recording pixel coordinates or DOM selectors. When a TPA redesigns its portal or a state updates its filing site, the recorded positions move and the bot breaks. The fix is to rebuild it, which is why so many of these projects quietly stall under their own maintenance weight.
Mediar reads the Windows accessibility tree instead, the same interface a screen reader uses. It finds a field by the name the application exposes, not by where it sits on screen. Here is the difference in concrete terms: the agent does not click coordinate (842, 318), it locates the element named Claim Number and types into it. When the portal restyles, the name survives.
That is the whole reason the accessibility-API approach exists. The agent watches a case manager or biller run the workflow once, learns the named fields it touches, then executes it. Because the contract is the field name and not the pixel, a portal redesign that would have sent a traditional bot back to the automation team leaves the Mediar run intact. The engine underneath is the open-source Terminator SDK, so a team that wants to extend or audit the workflows can.
Where Mediar is not the answer
If a step already has a clean API or sits on a stable, modern web app, you do not need an accessibility-tree agent for it; a normal integration or a conventional RPA bot will hold up fine there. Mediar’s edge is specifically the legacy, no-API, multi-portal desktop layer where pixel and selector bots cost the most to keep alive. It is also not a clinical decision tool: it does not decide medical necessity or write a care plan. It executes the repetitive data movement around those decisions.
And to be clear about the thing that started this page: if you needed a published, citable Triune Health Partners RPA case study to take to a committee, that artifact does not exist. What does exist is a real class of operation, workers’ comp case management and bill review, with a real automation problem, and a way to work the no-API screens at its center without rebuilding a bot every time a portal changes.
Bring your own bill-review or case management screen
Book a pilot and we will run one of your real no-API portals through Mediar, learning the named fields once, so you can see the self-healing behavior on a workflow you already do by hand.
Frequently asked
Frequently asked questions
Is there a UiPath, Automation Anywhere, or Blue Prism case study about Triune Health Partners?
No public one. As of June 2026, searching the three RPA vendors' case study libraries and the open web turns up no automation case study named after a company called Triune Health Partners. The name that does exist is TRIUNE Health Group, a workers' compensation medical case management firm in Lombard, Illinois, and it is not published as an RPA showcase. If you arrived expecting a named bot-deployment story with hours-saved numbers, that artifact does not exist publicly. What follows is the honest version: who the real company is, and what automating that kind of revenue cycle actually requires.
Who is TRIUNE Health Group and what do they do?
TRIUNE Health Group is a family-owned health care management company headquartered at 200 West 22nd Street in Lombard, Illinois, operating for more than 35 years. Their work is workers' compensation, not hospital billing: medical case management, nurse triage, catastrophic case management, utilization review, telephonic case management, vocational rehabilitation, personal injury protection, and life care planning. That distinction matters, because the revenue cycle behind workers' comp case management looks nothing like the hospital revenue cycle most RPA case studies describe.
Why is workers' comp revenue cycle different from hospital RCM?
Hospital RCM runs through one payer relationship per claim and a clearinghouse that speaks the 837 and 835 EDI standards. Workers' comp does not. A single bill can route to a third-party administrator, an employer, a state fund, and a bill-review vendor, each with its own portal and its own jurisdictional fee schedule. There is no universal clearinghouse that covers all of it, and many of the systems involved (case management platforms, bill-review tools, state filing portals) expose no API. So the work that should be automated is exactly the work that has no programmatic endpoint: a person reading one screen and typing into another.
Can UiPath, Automation Anywhere, or Blue Prism automate this?
They can drive the screens, but the no-API, multi-portal nature of workers' comp is where pixel-matched and selector-based bots are most brittle. Each TPA portal, each state filing system, and each bill-review tool has its own layout, and those layouts change. A bot recorded against coordinates or DOM selectors breaks on a redesign and goes back to the automation team to be rebuilt. That maintenance tax is the quiet reason many of these projects stall before they pay back. The generic RPA platforms are real tools; the friction is the brittle way they reach a screen that has no API behind it.
How does Mediar reach a screen with no API?
Mediar reads the Windows accessibility tree, the same interface a screen reader uses, so it locates fields by their exposed name rather than by pixel coordinates or DOM selectors. The agent types into a field called Claim Number or Date of Service, not into a rectangle at fixed coordinates. When a portal moves that field or restyles a button, the field is still named Claim Number, so the run keeps going. It watches a workflow once, then executes it. The engine underneath is the open-source Terminator SDK, so a team can extend the workflows themselves.
Is Mediar HIPAA compliant, and where does it run?
Mediar is SOC 2 Type II certified and HIPAA compliant, and deploys on-prem or in your own cloud. Because the agent runs on the same workstation or virtual desktop a case manager or biller uses, protected health information and claim data stay inside your environment. The agent operates the existing applications rather than exporting data to a separate service.
What does Mediar cost compared with enterprise RPA?
Pricing is $0.75 per minute of runtime with no per-seat licensing. There is a $10,000 turn-key program fee that converts to credits, so it is effectively prepaid usage. The contrast with enterprise RPA is the cost shape: per-bot licensing plus a months-long implementation and ongoing maintenance, versus runtime billing with the agent learning a workflow in days.
Related reading
Kaiser, Epic Resolute, and RPA claims automation: who does what
The same honest vendor-disambiguation, applied to hospital claims and Epic Resolute work queues.
Optometry revenue cycle automation, stage by stage
Where an optometry revenue cycle stalls when you try to automate it, and the cash-application layer with no API.
EMR billing workflow automation: the layer with no API
The denial and billing mechanics worked end to end through a no-API desktop client.
Comments (••)
Leave a comment to see what others are saying.Public and anonymous. No signup.