Nearly two-thirds (63%) of medical group leaders say their revenue cycle operations are leaning toward more automation in 2026, according to an Oct. 20, 2026, MGMA Stat poll. Another 26% report no major change, and 11% say their operations are shifting toward more manual work. The poll had 214 applicable responses. 

When Becker’s Hospital Review asked health system revenue cycle leaders about their biggest threats this fall, payer AI topped the list. Payers increasingly use automation to review claims, while provider organizations often lack the capacity to scrutinize those decisions at the same speed. 

In a Feb. 27, 2024, MGMA Stat poll, 62% of practice leaders said their organizations had automated 40% or less of revenue cycle operations.² That earlier poll measured how much RCM work had been automated. This week's responses offer a different perspective: where automation is expanding and where staff still handle work manually. 

Where practices are automating 

Among practice leaders moving toward more automation, coding was mentioned most often, followed by prior authorization, denials and appeals, and payment posting. Other respondents pointed less often to claim creation and submission, patient billing, eligibility verification and scheduling. Several answers listed whole chains, such as "claim creation, claim submission, payment posting, claim error rules." One answer was simply, "all — changed to Epic." Another leader was less sure the tools would deliver, describing an EHR with "so many workarounds that require manual interventions." 

Earlier polls show which RCM tasks practices have prioritized for automation. In a Jan. 27, 2026, MGMA Stat poll, automation was the most common answer (36%) when leaders named their biggest planned cost-cutting move, and respondents listed eligibility checks, claim status, denials workflows and prior authorization workflows among the work they planned to automate.³ Separately, an April 2026 poll found 33% of practices using AI for coding, charge review or both.⁴ 

Independent research on AI coding accuracy remains limited. A July 2026 Government Accountability Office review found few independent studies of AI coding accuracy and noted that cost and technical support needs may keep small practices from adopting these tools.⁵ 

At the MGMA 2026 Annual Conference, former CMS Administrator Seema Verma, now an Oracle Health executive, cautioned against stacking separate bolt-on tools, such as one AI agent for the billing team and another for physician coding, without a plan for how they share data.⁶ 

Where manual work is growing 

For the few practice leaders seeing a shift toward manual work, the top reasons given were payer complexity, tool gaps and staffing issues. One respondent said "AI and electronic sharing of files" were "not resulting in payment." Others pointed to "prior authorizations and AI audits" or simply "payer mischief." 

Payer requirements are a major source of manual work for physician practices. In MGMA's 2026 Regulatory Burden Report, 90% of practices said their prior authorization burden grew over the previous 12 months, and 40% said they employ three or more full-time administrative staff per physician for work such as prior authorization, denials and quality reporting.⁷ 

An approval does not always end the work. At a Sept. 29 Government Affairs town hall during the MGMA Annual Conference, an oncology practice leader described claims denied despite an approved authorization and coded as if the practice had erred, which left staff reviewing each claim to show the authorization was in place.⁸ The CMS WISeR model also introduced prior authorization for selected traditional Medicare services in six states in January 2026, which MGMA's Anders Gilberg faulted at the same session for running on third-party portals and fax rather than standardized electronic transactions.⁸ 

The difference in speed is especially apparent when practices appeal denied claims. Shawntea "Taya" Gordon, MBA, FACMPE, an MGMA consultant and CEO of Atlas & Perpetua Healthcare, told Physicians Practice at the conference that payers now review and deny claims faster than most practices can appeal them by hand, and that unmanaged denials and claims sitting in A/R are the leaks she finds most often.⁹ 

Downcoding can go unnoticed unless practices compare the CPT codes submitted with those ultimately paid on the same claim IDs.⁹ Automatic downcoding by Medicare Advantage plans is the fastest-growing complaint MGMA hears from members, and many practices do not know a plan has placed them under such a policy, MGMA's Madison Hynes told conference attendees.¹⁰ 

What keeps the mix steady 

Leaders who report no major change typically cited confidence in their current model or team; one respondent pointed to an experienced in-house billing team that was "beating the stats." Others cited budget constraints and/or reluctance or competing priorities ("We're slow to change and adopt more efficient methods."). One leader's confidence came with a caveat: "[I] don't want to lay off friends." Another respondent pointed the finger at new technologies "not working as expected." 

When an October 2025 MGMA Stat poll asked where groups would put their biggest new investment in 2026 budgets, 12% named revenue cycle, behind workforce (37%) and health IT (30%).¹¹ 

Capacity is a constraint alongside budget. Julia Rosen, MGMA's senior vice president of IT, told Medical Economics at the conference that many independent practices lack IT staff and are trying to stay current without the budget for licensed tools. Her advice for groups starting small: use the AI features already in the EHR or an enterprise AI tool approved for use with protected health information, with the necessary privacy safeguards and vendor agreements in place, and start in revenue cycle, where denial root-cause analysis on remittance files tends to pay back quickly.¹² 

Sorting practice work from payer work 

Some manual revenue cycle work comes from processes practices can improve themselves, including registration, eligibility checks, claim scrubbing and payment posting. Other work stems from payer requirements that practices cannot change on their own.  

Process redesign and automation can reduce that work, provided steps come out when a tool goes in. Shannon Cameron, who leads revenue cycle operations for Harvard Medical Faculty Physicians in Boston, told Becker's that automating a broken process repeats the same mistake across thousands of claims.¹ MGMA's January poll analysis suggested tracking touches per claim and per denial to confirm a tool actually removed work.³ 

The rest is generated upstream by payer prior authorization rules, Medicare Advantage downcoding policies and the WISeR model, and a practice cannot automate it away alone. One practical response is better documentation. A payer-by-payer log of denials on approved authorizations and downcoded claims gives leaders evidence for contract talks and supplies the examples MGMA carries to congressional offices.⁸^,¹⁰ Such records may also help practices challenge AI-assisted medical-necessity denials under new state laws, where applicable. These protections vary by state and generally do not cover self-funded employer plans or Medicare Advantage.¹³ 

This week's comments show how the two trends overlap. Practices moving toward automation most often cited coding, prior auth, denials management and payment posting. Payer complexity led the reasons for increased manual work. Prior auth appeared on both sides: a target for automation and a source of follow-up, appeals and audits that still demand staff time. 

For independent groups, much of what comes next depends on an EHR vendor. Payers covered by the CMS Interoperability and Prior Authorization final rule must generally have prior authorization application programming interfaces (APIs) in production beginning Jan. 1, 2027, yet Gilberg said few members MGMA staff asked at the conference had heard from their EHR vendors about it.¹⁴ That leaves practices a concrete question for EHR and PM vendors this quarter: which payers' prior authorization APIs will connect to the practice's system, when and at what cost? 

Notes 

1. Dyrda L. "10 hospital revenue cycle risks and what automation makes worse." Becker's Hospital Review. Oct. 2, 2026. Available from: https://www.beckershospitalreview.com/finance/revenue-cycle-management/10-hospital-revenue-cycle-risks-and-what-automation-makes-worse/  
2. Harrop C. "Steps toward revenue cycle automation vary across medical groups." MGMA. Feb. 28, 2024. Available from: https://www.mgma.com/mgma-stat/steps-toward-revenue-cycle-automation-vary-across-medical-groups  
3. Harrop C. "Automation, process fixes top list of cost-cutting moves for medical practices in 2026." MGMA. Jan. 28, 2026. Available from: https://www.mgma.com/mgma-stat/top-cost-cutting-moves-for-medical-practices-in-2026  
4. Harrop C. "Before you buy: Evaluating AI for coding and charge review in your practice." MGMA Stat. April 8, 2026. Available from: https://www.mgma.com/mgma-stat/before-you-buy-evaluating-ai-for-coding-and-charge-review-in-your-practice  
5. U.S. Government Accountability Office. "Science & Tech Spotlight: AI for Medical Notes and Coding." GAO-26-109116. July 16, 2026. Available from: https://www.gao.gov/products/gao-26-109116  
6. Payerchin R. "People, process, technology: AI success takes more than software, former CMS chief says." Physicians Practice. Sept. 30, 2026. 
7. MGMA Government Affairs. "MGMA 2026 Regulatory Burden Report." April 9, 2026. 
8. Littrell A. "Practice leaders tell MGMA an approved prior authorization is no guarantee of payment." Physicians Practice. Sept. 30, 2026. Available from: https://www.physicianspractice.com/view/people-process-technology-ai-success-takes-more-than-software-former-cms-chief-says  
9. Reynolds KA. "Payers deny claims faster than practices can appeal, consultant Taya Gordon says." Physicians Practice. Oct. 8, 2026. Available from: https://www.physicianspractice.com/view/payers-deny-claims-faster-than-practices-can-appeal-taya-gordon  
10. Littrell A. "MGMA details what practices face in January, from Medicaid work requirements to remote monitoring rules." Physicians Practice. Sept. 28, 2026. Available from: https://www.physicianspractice.com/view/mgma-details-what-practices-face-in-january-from-medicaid-work-requirements-to-remote-monitoring-rules  
11. Harrop C. "Where medical groups are putting new dollars in 2026 budgets." MGMA. Oct. 29, 2025. Available from: https://www.mgma.com/mgma-stat/where-medical-groups-are-putting-new-dollars-in-2026  
12. Payerchin R. "Start with revenue cycle: MGMA's IT chief on where practices should begin with AI." Medical Economics. Oct. 6, 2026. Available from: https://www.medicaleconomics.com/view/start-with-revenue-cycle-mgma-s-it-chief-on-where-practices-should-begin-with-ai  
13. Reynolds KA. "State AI insurance laws give practices new grounds to fight denials and downcodes." Physicians Practice. Oct. 16, 2026. Available from: https://www.physicianspractice.com/view/state-ai-insurance-laws-give-practices-new-grounds-to-fight-denials-and-downcodes  
14. Reynolds KA. "MGMA warns payers against backing out of the Jan. 1 prior authorization deadline." Physicians Practice. Oct. 1, 2026. Available from: https://www.physicianspractice.com/view/mgma-warns-payers-prior-authorization-api-deadline-2027