I had never called my own practice at 9 p.m. Most surgeons I know have not either. Our patients do it constantly. That gap bothered me enough to start calling practices after hours the way a patient would. What I heard was rarely a technology failure. It was a set of design decisions nobody remembers making: who answers, what they collect, and who decides whether the call can wait until morning.
The call you are judged on is not the one you planned for
Daytime volume is a staffing question. You can watch the queue, count the abandons and add a person. After hours behaves differently, because the volume is low and the variance is enormous. Most nights nothing happens. Then a patient 10 days out from a craniotomy calls at 11 p.m. because she is more confused than usual and has vomited twice, or a fusion patient calls because a leg has gone weak, and the entire arrangement is judged on that one call.
What is clinical and what is clerical
This is where after-hours coverage usually goes wrong, and it goes wrong in both directions.
Deciding whether a new neurologic deficit needs to be seen tonight is clinical. Deciding whether a post-op wound question can wait until morning is clinical. Deciding which surgeon covers which service on a Saturday is not. Collecting the patient’s name, date of birth, surgery date and a working callback number is not. Reaching the right person is not.
Practices collapse these together in one of two ways. The common one is a service that takes a message and pages whoever is on the list, which treats every call as clerical and hands the clinical judgment to a surgeon reading 20 words with no chart in front of him. The rarer one runs the other way: the surgeon is woken and then spends 10 minutes working out who the patient is and what operation they had.
The split worth holding is that the clerical half should be complete and accurate before any clinician is involved, and the clinical half should never be decided by someone without the criteria to decide it.
Which raises the question of whose criteria.
Write down the criteria you already use
Most practices have escalation criteria. Few have them written down. They live in the head of whoever takes the most call, and they get applied differently depending on who answers the phone.
Writing them down costs nothing and is the highest-yield item on this list. Not a triage protocol in the clinical sense. Just your practice’s own answer to three questions: what wakes me up, what waits until morning, and what the person answering has to collect before either. Post-op day, procedure, symptom, red flags by service. A page or two.
Once it exists, three things become possible that were not before. Whoever answers can apply it the same way twice. You can audit whether it was applied. And you can change it deliberately instead of by attrition.
Measure your own line before you change it
Call your own main number at 9 p.m., and again on a Sunday afternoon. Do it twice each time: once as a new patient trying to book, once describing a post-op symptom. Write down what happened, how long it took, and whether someone in pain would have stayed on the line.
Most practices find at least one thing they did not know was true. A voicemail box that is full. A tree that loops back on itself. A service that takes 4 minutes to reach a person and then offers only a callback.
That test takes 20 minutes and costs nothing. It is the only step here that produces a number specific to your practice, and every decision after it is better for having one.
What I would not claim
Automation does not resolve the clinical question. Whatever answers the phone, somebody still has to decide what wakes the surgeon, and that decision belongs to the practice rather than to a vendor or a piece of software. The gains available after hours sit almost entirely in the clerical half: answering at all, collecting the right information, and routing to the right person with the record intact.
The clinical half is yours. It is worth writing down before anyone automates around it.
By Matt Fisher, VP of Operations, Curae; president, HFMA Georgia.
The next revenue cycle risk for health systems may not arrive as a denied claim or an unpaid statement. It may first appear as work that is harder to see, such as a missed health insurance premium renewal, a patient who can no longer afford their marketplace or employer-sponsored health plan , or an out-of-pocket responsibility that exceeds what the patient can realistically pay.
That hidden work matters because it reaches the revenue cycle before a balance ever ages. Coverage and affordability issues often begin upstream during scheduling, when staff are verifying coverage, determining expected out-of-pocket costs, and discussing financial options with patients. By the time an account reaches the back end, the balance may have originated weeks earlier due to an enrollment issue, an affordability challenge, or a patient who did not understand their financial responsibility.
Medicaid, the Affordable Care Act, and broader affordability changes are creating a continuous cycle of coverage disruption that will play out over the next two years. Each policy shift creates another point where patients can lose coverage, misunderstand their benefits, or face costs they cannot realistically afford. ACA premiums were projected to rise by a median 15% in 2026, based on a KFF analysis of preliminary filings across 19 states and Washington, D.C.
New Medicaid work requirements will also require some enrollees to document 80 hours of work or qualifying activities each month, creating another point where eligible patients may lose coverage because of administrative barriers rather than a lack of need for care. For many of those patients, the challenge does not end with losing Medicaid. Transitioning to Marketplace coverage may mean facing premiums they cannot afford, creating a separate affordability issue where strategies such as premium sponsorship can help eligible patients maintain coverage and access to care.
CFOs and revenue cycle leaders need a fuller view of coverage risk than a yes-or-no insurance check can provide. A patient may have coverage at registration but still be at risk of losing it, misunderstanding it, or facing a balance that will move quickly into bad debt. Patients may remain insured while still being unable to afford care because of rising deductibles, premium contributions, or cost-sharing requirements, creating the rise of the “insured but unaffordable” patient.
Eligibility Friction is Becoming a Cash Flow Problem
Eligibility friction creates work across health systems long before an account reaches collections. Outdated coverage can affect registration, missing documentation can lead to denials, and a plan change can leave patients confused by an unexpected deductible. Patient collections teams may eventually receive balances that would have been better routed earlier to financial assistance, charity care, enrollment support, or a medical payment plan.
The downstream risk is larger because many households have little room to absorb medical bills. Medical collections accounted for 57% of all collections tradelines on consumer credit reports from 2018 to 2022, according to the Consumer Financial Protection Bureau. Once a balance reaches collections, the health system has usually lost the best chance to preserve coverage, recover a claim, or match the patient with an appropriate payment path.
Patient Access Needs Earlier Financial Signals
A cleaner front-end process can prevent some accounts from taking the long route through denials, statements, and collections. Eligibility data, prior balances, high-deductible exposure, and propensity to pay can help revenue cycle teams identify when a patient may need support before the bill is finalized.
That support will not look the same for every account. A patient coming off Medicaid may need help with enrollment or marketplace guidance. A patient with a correctable coverage issue may need a claims review before the balance shifts to patient responsibility. A patient with a valid balance may need health service financing, an interest-free medical loan, a medical line of credit, or a medical payment plan that fits the household’s actual budget.
With the right financial context, patient access can help identify risk earlier and direct patients toward enrollment help, financial assistance, or a payment option before the account moves further downstream.
Financing Should Fit the Patient’s Situation
Patient financing programs will become more important as coverage instability leaves more patients with larger balances, but financing works best when it follows a clear review of the account. Health systems first need to determine whether the patient has an eligibility issue, a claim issue, a financial assistance need, or a true patient responsibility balance. When financing is appropriate, options may include shorter-term in-house payment arrangements for smaller balances and longer-term zero-interest financing for larger patient responsibilities, depending on the organization’s approach and the patient’s financial situation.
When financing is appropriate, revenue cycle leaders should evaluate patient financing companies and services based on access, compliance, patient clarity, and provider risk. The best options should be accessible to patients with limited savings, provide funds quickly, comply with consumer credit requirements, and support a positive patient financial experience. Programs built mainly around high-credit patients may improve the patient financial experience for a narrow group, but they will not address many of the balances most likely to become bad debt.
Non-recourse patient financing may be part of the evaluation because it can reduce provider exposure if the patient does not repay. Healthcare payment technology should also connect financing to patient access, billing, and follow-up, ensuring payment options fit into the broader financial access process.
Health Systems Need a Two-Year Operating Plan
The next two years are likely to bring repeated coverage changes rather than one clean disruption. Health systems will need a way to catch coverage issues early enough to keep patients from drifting into avoidable balances.
That work will affect healthcare finances well beyond collections. A missed renewal or unaffordable plan can change payer mix, strain cash flow, and increase uncompensated care before the account ever reaches the back end. Pushing harder on collections may recover some balances, but health systems cannot collect their way out of growing coverage instability and affordability challenges. The larger opportunity is identifying financial risk earlier, preserving coverage whenever possible, and connecting patients with the resources they need before balances become bad debt.
Health systems that prepare now will be better positioned to protect margins while helping patients maintain access to care. As coverage disruption continues over the next two years, organizations that treat financial access as an ongoing operational function, not simply a collections function, will create better outcomes for both their organizations and the patients they serve.
By Thomas Thatapudi, Chief Information Officer, AGS Health.
In healthcare revenue cycle management (RCM), we’ve long relied on systems that process rules, not judgment. Robotic Process Automation (RPA) has been beneficial, automating repetitive, high-volume tasks such as claim status checks and data entry. However, its limitations are increasingly apparent. Today’s revenue cycle challenges demand more than just speed and efficiency; they require adaptability, context, and intelligent decision-making.
That’s where agentic AI comes in.
Agentic AI represents a next-generation approach to automation—one that mimics how humans think, make decisions, and interact with systems and people. Unlike RPA, which follows strict, predefined scripts, agentic AI models operate like autonomous agents. They’re context-aware, goal-oriented, and capable of reasoning across complex workflows. For revenue cycle teams under pressure from rising denials, staffing shortages, and shrinking margins, this kind of intelligence isn’t just nice to have—it’s becoming essential.
What Makes Agentic AI Different?
The simplest way to explain agentic AI is to compare it to a seasoned team member—one who not only knows how to complete a task but also when to escalate, adapt, or reprioritize based on changing circumstances. Agentic systems can:
Interpret and act on real-time data from multiple systems
Make decisions without human intervention
Learn from patterns and improve over time
Collaborate with human team members when needed
In practical terms, this means AI can now triage claims, initiate and complete payer calls, route work dynamically, or even autonomously document and code encounters—all with human-like logic and consistency.
Why This Matters for RCM
Healthcare RCM is a perfect candidate for this type of automation because it sits at the intersection of structure and unpredictability. Processes are highly regulated, but real-world conditions vary constantly. Consider these examples:
Accounts receivable: Agentic AI can identify which claims require expert attention and which can be resolved through automation, ensuring staff spend their time where it’s most needed.
Insurance follow-ups: AI agents can navigate payer phone trees, wait on hold, retrieve claim information, and even update the EHR, without tying up human resources.
Denial management: Instead of flagging a denied claim for review, an agentic system can analyze the denial reason, check documentation, and suggest or initiate corrective actions.
These aren’t distant possibilities—they’re already being piloted and implemented in real-world environments.
The Human + Agentic AI Model
It’s important to note that agentic AI is not about replacing people—it’s about augmenting them. The most effective models combine human oversight with AI execution:
Human experts handle edge cases, complex judgment calls, or relationship-driven tasks.
AI agents handle high-volume, rule-governed, or low-dollar work with consistency and speed.
This hybrid approach doesn’t just improve throughput; it also enhances job satisfaction for teams that no longer spend their days on tedious follow-ups or simple reconciliations.
Getting Started with Agentic AI
For organizations beginning to explore this space, here are a few guiding steps:
Consolidate and clean your data: Fragmented data across EHRs, billing systems, and vendor platforms limits AI effectiveness. Start by creating interoperable, governed data environments.
Identify high-ROI use cases: Look for repeatable processes with moderate complexity and clear financial upside, like denial prediction, prior authorization automation, or A/R follow-ups.
Experiment with short feedback loops: Choose pilots where you can quickly assess ROI and adjust based on results. Don’t aim for perfection—aim for momentum.
Build trust through transparency: Ensure your AI systems are auditable and explainable, especially when financial decisions are being made autonomously.
A Path to Sustainable Margins
Every healthcare leader is being asked to do more with less: deliver care, navigate compliance, and protect financial performance. Agentic AI offers a path forward, not as a magic bullet, but as a powerful tool for reclaiming time, improving accuracy, and aligning resources where they have the most impact.
It’s still early days for agentic AI in healthcare RCM, but the direction is clear. With the right balance of vision and pragmatism, revenue cycle leaders can unlock a new level of operational intelligence and move closer to sustainable, value-driven performance.
Financial pressure has become a persistent reality for hospitals and health systems. Hospital expenses increased 17.5% between 2019 and 2022, and more than half of hospitals ended 2022 operating at a loss. Although performance has improved from those historic lows, the underlying cost pressures have not gone away. In 2025, total hospital expenses grew another 7.5%, more than twice the rate of growth in hospital prices, with increases across workforce, supplies, and drugs.
These pressures make disciplined expense management increasingly important. Yet hospital operators have limited flexibility across many of their largest expense categories. About 60% of total hospital expenses in 2025 went to the workforce needed to provide around-the-clock care, while supplies and drugs also continued to rise. Hospitals cannot simply reduce these resources without considering the potential consequences for capacity, quality, and patient care.
That makes indirect and outsourced spend – commonly referred to as purchased services – an increasingly important opportunity. Purchased services encompass a broad range of essential functions, from facilities management and information technology to revenue cycle, clinical engineering, food services, security, and other outsourced operations. Purchased services expense per calendar day increased approximately 9% year over year in September 2025, according to Kaufman Hall hospital performance data.
While cost pressures in these categories continue to rise, so does the opportunity to manage them more effectively. The challenge is that hospital operators responsible for departmental expense budgets often lack the visibility, data and tools needed to define a clear path to savings.
A clearer view of purchased services spending across departments, vendors, contracts, invoices, and general ledger accounts is therefore critical to enabling operators to support the organization’s budget targets. Without that visibility, expenses may already be incurred or contractually committed before operators have an opportunity to effectively plan and manage them. With it, health systems can identify pricing differences, coordinate vendor relationships, strengthen negotiations and prioritize savings opportunities without compromising the services on which their organizations depend.
Hospitals Lack Visibility into Their Purchased Services Spend
One reason hospital operators are challenged to meet expense budget targets for indirect and outsourced services is limited visibility into current spending.
Purchased services data is frequently fragmented across departments, contracts, accounts payable systems and enterprise resource planning platforms. Individual departments may independently manage relationships with the same supplier or purchase similar services from different suppliers. At the same time, purchased services invoices may be applied inconsistently across multiple general ledger codes. Large “catch-all” expense accounts and unoptimized ERP data structures can make it difficult for operators to determine precisely where spending is occurring and how it compares across departments or facilities.
The result is not simply a reporting problem. It limits an organization’s ability to actively manage spending before decisions become financial commitments.
For example, one department might negotiate favorable pricing for a particular service while another department contracts separately for the same or a similar service at a higher rate. A contract might also automatically renew with limited scrutiny while annual price escalators continue to increase costs. These differences can weaken the health system’s negotiating leverage by fragmenting purchasing volume and obscuring the organization’s total relationship with a supplier.
Classification compounds the problem. Without a standardized taxonomy for purchased services, such as UNSPSC, similar expenses may be categorized differently across facilities or assigned to broad general ledger accounts. Operators are then left without consistent, actionable insight into the line-item expenses they are expected to manage.
Inconsistent, one-off approaches to purchased services management make it difficult for operators to consistently achieve budgeted expense targets. Health systems instead need a unified view of spending and a repeatable process for identifying, capturing, and sustaining savings.
A Framework for Better Spend Intelligence
Health systems need a framework that increases visibility into purchased-services spending and establishes a consistent process for organizing, comparing, and acting on information across the enterprise. Four capabilities are particularly important:
Consolidate spend data. Create a unified, drill-down view of purchased services spending across departments, vendors, contracts, invoices, and general ledger accounts. Connecting these data sources helps operators identify duplicate agreements, fragmented supplier relationships, unexpected pricing differences, and other opportunities that may be difficult to see within individual systems.
Standardize service classifications. Apply a consistent taxonomy to purchased services so similar expenses are categorized the same way across departments and facilities. Accurate classification gives operators a more reliable view of what they are spending and enables meaningful comparisons across the health system.
Benchmark performance. Compare spending, utilization, and pricing across facilities within an integrated delivery network and, where appropriate, against external market benchmarks. Benchmarking can help operators identify unexpected variances, determine where costs may be above market and enter supplier negotiations with stronger supporting data.
Prioritize high-impact opportunities. Focus first on categories where the combination of spend, pricing variance and addressable opportunity offers the greatest potential financial impact. Prioritization allows operators to direct limited resources toward initiatives most likely to support near-term budget objectives.
This framework provides a starting point, but identifying an opportunity is not the same as capturing or sustaining it. Hospitals must also consider service quality, operational continuity, contractual requirements, and staff workload. The lowest-priced option is not necessarily the option that delivers the greatest value.
A sustainable purchased services strategy therefore requires more than periodic sourcing exercises. It requires an operating model that continually connects financial insight with operational decision-making.
Sustaining Savings Beyond Initial Analysis
Purchased services savings can deteriorate quickly if they are treated as one-time initiatives.
Centralizing spend information, for example, creates value only if the data remains current and operators continue to use it. Supplier consolidation can improve purchasing leverage, but reducing the number of vendors alone will not sustain savings without ongoing visibility into pricing, utilization, performance, contract terms, and changing service requirements.
Technology can help institutionalize that discipline. Analytics platforms can continuously organize and analyze purchased services data, identify spending patterns and variances, apply relevant benchmarks, and provide operators with information they can use when managing budgets or negotiating with suppliers. Just as importantly, a shared data foundation can keep finance, supply chain and operational leaders working from the same view of the organization’s spending.
Cross-functional coordination is essential. Finance can establish budget expectations and provide enterprise-level financial visibility. Supply chain can bring sourcing expertise, contracting discipline and market intelligence. Operators can assess service requirements, utilization, and performance within the departments they manage.
Each function has a different role, but sustainable savings depend on those roles working together rather than operating in separate silos.
For some health systems, establishing that capability may also require external expertise to supplement internal teams, particularly across specialized purchased services categories where market pricing, contract structures and supplier dynamics can be difficult to assess.
Cost Containment Starts with Clearer Spend Insights
Financial pressures on hospitals are unlikely to disappear. Rising labor, drug, supply, and administrative expenses continue to challenge organizations already operating within narrow margins. At the same time, many of the largest expense categories offer limited flexibility without potentially affecting access, staffing, or patient care.
Purchased services are different.
These expenses support essential hospital operations, but fragmented contracts, inconsistent categorization, limited visibility, and decentralized purchasing can leave significant opportunities unmanaged. Improving performance does not require indiscriminate reductions in services. It requires giving hospital operators the information and processes they need to make better decisions about the expenses they already control.
Health systems that create a unified view of purchased services spend can more readily identify pricing disparities, understand supplier relationships, strengthen negotiating leverage and focus resources on opportunities with the greatest financial impact. They can also evaluate those opportunities in the context of service quality, operational requirements, and patient experience.
In an environment where every margin point matters, sustainable cost containment depends on transparency, accountability, and discipline. By turning purchased services from a fragmented expense category into a continuously managed source of spend intelligence, hospitals can give operators a clearer path toward meeting budget targets – and create a more durable foundation for financial performance.
The US healthcare system is coming up against a variety of challenges, but one of the most crucial is meeting the needs of our aging population. As the care requirements of the long-tail Baby Boomer generation shifts, their needs are predicted to surpass the capacity of hospitals across the country in short order.
In what is being called “The 2030 Problem,” that milestone year will see the entirety of this demographic exceed the age of 65, equating to one out of every five Americans. The US Census Bureau estimates that more than 11,000 people turn 65 each day. In an analysis by Dr. Richard Leuchter, University of California, Los Angeles, he states that without significant intervention, hospitals could reach a critical 85% occupancy by 2032. This represents a dangerous threshold, since CDC modeling data has revealed that when ICU occupancy hits 75% nationwide, an estimated 12,000 excess deaths will occur within the following two-week period.
Connectivity as a Critical Healthcare Infrastructure
It will literally be impossible for hospitals to physically expand the amount of staffed, facility-based beds required to stave-off a crisis by this token. Hospitals have no choice but to expand their care virtually, modernizing their infrastructures to accommodate digital-first clinical operations including remote monitoring and virtual in-home care.
Thousands of facilities are currently scrambling to figure out how to accomplish this, because modernization of technology can be expensive, and networks that can accommodate both internal connectivity and mobile or remote care must be both extremely robust, expansive, and secure. This means that Wi-Fi and cellular have become not just a nicety for high-speed Internet, but more of a table-stakes utility, as vital as electricity and heat, in order to keep hospitals serving the needs of the public into the future.
According to the American Telemedicine Association (ATA) the roadmap to an effective digital-first hospital contains six major goals, comprised of both immediate priorities and long-term objectives. Literally half of these involve upgrades to broadband and cloud-based digital infrastructures that seamlessly integrate patient data, including the integration of edge computing to support AI “to ensure continuous, high-speed connectivity over all care settings.”
As networking capabilities become more sophisticated, hospital administrators have found that a single infrastructure (e.g., either 5G/6G or Wi-Fi) is not satisfactory in addressing the needs of both internal connectivity and mobility. A combination of top-flight 5G/6G and Wi-Fi networks are required to address the demands of both existing internal hospital systems and allow for the expansion of the network into mobile and virtual settings. This is the key to increasing capacity without having to physically build new spaces that will house the aging patients of the future.
A converged cellular/LAN/WAN connectivity network addresses a multitude of issues currently affecting healthcare organizations across the country.
Improved clinical communication: High-end connectivity will dramatically improve internal communication and seamless transfer of data between different areas of a healthcare campus. Practitioners are consistently traversing hospital buildings, going from room to room, floor to floor, and even campus to campus. Too often, ill-designed Wi-Fi networks cause doctors and nurses to experience data drops as they enter an elevator or proceed to a patient’s room in a far-flung wing. In those instances, digital information can altogether disappear from their devices. This leads to practitioners taking work home with them, aggravating their workload and contributing to staff burn-out.
Shift to Always-On Care: Trends have been accelerating towards telehealth installations, use of mobile clinicians and hybrid or work-from-home scenarios for overburdened practitioners, and emerging remote monitoring of patients. This is especially useful in addressing the “Aging in Place” trend to deliver more care for seniors in their home environments as they continue to age, as opposed to requiring them to frequent a facility.
Isolated recovery environments: Hospitals have a vital need for data resiliency in the cloud, to ward off any network failures through either natural disaster, or the intervention of malicious hackers. It hasn’t been beyond the realm of possibility that a hospital system gets taken for ransom or is knocked off-line by a weather emergency—denying the ability for doctors and nurses to prescribe medication, review medical records, or utilize connected equipment. This is simply not an acceptable scenario.
Required Continuity: Hospitals tend to plan for power outages, but not network outages. Yet delivery of care now depends heavily on connectivity uptime. A 5G/6G network with a major carrier can serve as a resiliency network, creating an active recovery environment. An intelligently designed fail-safe system can allow data traffic to immediately divert to a resilient network layer, empowering a hospital to continue functioning, providing contiguous care as normal until the main network is restored.
Continuity of care is a key initiative of the Department of Homeland Security. DHS and FEMA have been inconspicuously reaching out to healthcare systems, compelling them to develop a continuity of operations plan (COOP) to ensure the viability of essential hospital functionality during emergencies or cyber-attacks. These functions include operations, patient care, and staffing. Much of the recovery initiatives recommended by government agencies and cybersecurity companies involve the ability to provide a seamless failover connectivity layer that explicitly connects to network back-up throughout a hospital campus.
This includes all networks associated with the facility. The truth is, bad actors are out there, looking to disrupt networks and steal data. And even less predictably, bad circumstances like weather emergencies will also always present a threat. In the event of a catastrophe—whether a terrorist attack, a climate-related disaster, or even a mundane ransomware issue—the DHS is demanding that healthcare companies have network resiliency plans in place.
“The future of healthcare is seamlessly connected care,” stated Rachelle Longo, AVP, Telemedicine from Ochsner Health, in an article for the ATA Center of Digital Excellence (American Telemedicine Association). “Those who invest in modern infrastructure today will define the next era of patient-centered, technology-driven care.”
Leveraging AI to Create Cost Defrayment
We know that modernization of a healthcare network is far easier said than done, however. Revamping a connectivity infrastructure has typically been a hugely complicated and costly undertaking, requiring detailed knowledge of telecommunications and requiring the cooperation of a multitude of technology providers and services management.
This is why new models such as working with an aggregator with knowledge of the telecom field and utilizing granular, AI-based cost-analyses to identify areas of savings that can be applied to defray costs are gaining ground. Such methods warrant an entire additional discussion, but consider that AI capabilities have never been powerful enough in the past to accommodate detailed enough analysis of healthcare organization expenditures to identify large enough areas of savings to help fund modernization.
So in this new era of high-performance cellular-Wi-Fi convergence, combined with the accelerated capabilities of AI, the healthcare industry can foresee a path to drive clinical operations forward that is not only technologically achievable, but also financially feasible.
The healthcare industry has invested billions in clinical AI, electronic health records, and back-office automation. Machine learning reads radiology scans. Predictive models flag at-risk patients. Robotic process automation handles insurance claims. And yet, most health systems have completely overlooked the single interaction that determines whether a patient receives care at all: the phone call.
For the vast majority of patients, the journey to treatment doesn’t start with a click. It starts with a call. They pick up the phone to schedule an appointment, ask a question about symptoms, or follow up on a referral. What happens in that moment, whether someone answers, how long the caller waits, whether the conversation leads to a booked visit, has an outsized impact on both patient outcomes and practice viability. And right now, that moment is failing at scale.
The Scope of the Problem
The numbers are stark. Research shows that medical and dental practices miss an average of 35% of incoming phone calls, with some offices hitting rates as high as 68%. In dentistry alone, one out of every five new patient calls goes unanswered. When patients do get through, 34% hang up after just two minutes on hold, and 85% won’t call back if their first attempt goes unanswered.
Each of those abandoned calls carries real consequences. A single missed new patient call in a dental practice represents roughly $850 in immediate revenue, with a lifetime patient value that can reach $8,000. Across the industry, the average practice loses between $100,000 and $150,000 annually from missed calls alone. Scale that across thousands of practices nationwide and the aggregate revenue loss reaches well into the billions.
But the financial toll is only part of the story. Behind every unanswered call is a patient who needed care and didn’t receive it. Someone whose cavity worsened, whose chronic condition went unmanaged, or whose referral expired before they could be seen. In a healthcare system already struggling with access, the front desk has quietly become one of the biggest bottlenecks to treatment.
Why the Front Desk Is Breaking
The root cause isn’t a lack of effort by front desk teams. It’s a structural mismatch between the volume and complexity of patient calls and the resources available to handle them. A 2026 survey found that 64% of healthcare providers say staffing shortages directly reduce patient access, up from 57% just a year earlier. One-third of practices report difficulty hiring administrative and front desk personnel.
At the same time, the demands on those staff members have intensified. A front desk employee today is expected to answer calls, check patients in, verify insurance, manage the schedule, and handle follow-ups, often all at once. When the phone rings during a busy check-in, the call goes to voicemail. When a new patient calls with questions about insurance or procedure costs, the conversation may take ten minutes that the staff member simply doesn’t have. The result isn’t poor service by choice. It’s an impossible workload.
A New Layer of the Technology Stack
This is where AI can make its most immediate and measurable impact in healthcare. Not in replacing clinicians, but in making sure patients actually reach them. A new category of AI-powered communication tools is emerging that sits at the front door of the practice, analyzing phone, chat, SMS, and online scheduling interactions in real time, coaching staff on how to convert inquiries into appointments, and identifying patterns that reveal why patients are falling through the cracks.
These solutions work at the intersection of natural language processing, call analytics, and workflow automation. They can detect when a caller is a new patient, flag calls where the staff member missed a scheduling opportunity, and surface trends like a spike in calls about a particular procedure or a recurring bottleneck at a specific time of day. Some platforms go further, offering AI-driven call handling that can answer routine questions, book appointments, and route complex inquiries to the right person so that no call goes unanswered even during peak hours.
What makes this category different from traditional call center software is the focus on outcomes, not just efficiency. The goal isn’t to reduce average handle time. It’s to increase the number of patients who successfully access care. That distinction matters because it aligns the technology with the mission of the practice: better patient outcomes. The old call center metrics provided by telephony or call center software providers are simply not actionable enough to drive meaningful change.
The Access-to-Care Imperative
More than 81% of healthcare leaders acknowledge that staffing-driven delays are creating substantial barriers to care, leading to longer waits for appointments and reduced access to screenings, diagnostics, and preventive services. When patients can’t get timely appointments, conditions deteriorate, emergency visits increase, and the overall cost of care rises.
AI-powered front desk technology addresses this problem at its source. By making sure every patient call is answered, analyzed, and acted upon, practices can recapture the patients they’re currently losing, without hiring additional staff they can’t find. The financial impact is significant: reducing missed calls by even a modest percentage can recover hundreds of thousands of dollars in annual revenue for a single practice. But the greater impact is measured in patients who receive the care they sought, when they sought it.
Looking Ahead
Healthcare has rightly invested in AI for diagnostics, treatment planning, and clinical decision support. Those are critical applications. But the industry has to recognize that none of those investments matter if patients can’t get through the front door.
Today, the phone call remains the primary point of entry for most patients, and it’s the point where the most patients are lost. But the front door is getting wider. SMS, web chat, and online scheduling are becoming increasingly important channels, and each one presents the same fundamental challenge: making sure every interaction leads to the right outcome for the patient. The practices that thrive will be the ones that deploy AI not only to interact with patients across these channels, but to monitor the quality of every single interaction, identifying where the patient journey breaks down and intervening before that patient is lost.
Enhancing the patient journey should be the goal of every AI technology in healthcare. AI won’t replace the empathy and judgment of a skilled front desk team. But it can give that team the tools, insights, and support to do what they do best: connect patients with the care they need. As the CEO of one of the first AI companies in healthcare, I take that obligation as a solemn one.
Most telehealth companies sell a medication with a subscription wrapped around it. A questionnaire. A prescription. A box that shows up.
Mochi Health works differently. The company connects patients, providers, and independent pharmacies on one platform. It’s the structure of that connection, not any single medication moving through it, that Mochi is actually selling.
That distinction matters more than it sounds. A prescription pipeline can be copied by any competitor with a similar drug supply.
A working three-sided marketplace, where each side has a reason to keep showing up, is considerably harder to replicate, and it’s the reason Mochi frames its business this way rather than as a weight-loss app with a pharmacy attached.
The Prescription Pipeline Most Telehealth Runs On
The standard direct-to-consumer model is a straight line. A patient fills out a form. An algorithm or a rushed provider approves a prescription. A single contracted pharmacy ships it.
The patient never chooses the pharmacy and rarely keeps the same provider twice. This model works fine for a single, simple request. It breaks down the moment a patient needs anything ongoing.
The Three Sides, Explained
Mochi’s marketplace gives each of the three parties something a straight pipeline doesn’t. Patients choose their own provider from a roster of more than 400 board-certified physicians, nurse practitioners, and dietitians, and they keep that same provider as their care continues, rather than being routed to whoever’s available.
Getting started is a short eligibility survey, then a provider match, and from there unlimited ongoing access to that care team rather than a single one-off visit.
Providers get the other half of the arrangement. The ability to practice without being boxed into a single condition or a script written by someone else’s protocol, building an actual panel of patients they know over time.
And independent pharmacies plug into Mochi’s software as fulfillment partners that patients can see and compare, competing for that business on service and price rather than being handed volume by default. None of the three sides is just a vendor. Each one has a reason to stay.
Why the Structure Is the Product, Not the Medication
The medication itself is close to a commodity; multiple pharmacies can produce comparable versions of the same GLP-1 or dermatology treatment. What isn’t commoditized is the structure that connects a specific patient to a provider who knows them and a pharmacy that’s been vetted and is competing to serve them well.
Mochi has described its role plainly, as “a software-enabled marketplace that’s making pharmacies better at what they do, which results in better patient outcomes.” This is a description of infrastructure, not of a drug.
That framing is also why Mochi’s pricing looks the way it does. A flat membership starting at $39 for the first month and $79 a month after, with no hidden fees, and medication priced and shown separately rather than bundled into an opaque total.
What This Means Day to Day for a Patient
In practice, a Mochi member gets an in-app care team, messaging with their provider, and 24/7 support that persists across their relationship, plus access to more than 9 treatment areas, from weight management to skincare, mental health, and menopause, without switching platforms or providers each time a new need comes up.
The company reports members averaging meaningful, sustained progress over the course of a year of continued care, the kind of outcome a one-off prescription pipeline isn’t structured to produce, because it was never designed to keep anyone around long enough to get there.
The Marketplace Is the Product
Most of the category still markets itself around the medication: whichever drug is trending, whichever price point undercuts a competitor this quarter.
Mochi’s bet is that the medication is the least differentiated part of the business and the marketplace connecting patient, provider, and pharmacy is the part actually worth building.
More than 500,000 patients choosing their own provider and pharmacy inside that structure is the evidence that the bet is working, and it’s the reason Mochi Health describes itself first as a marketplace, and only second as anywhere you can get a prescription filled.
Healthcare organizations can no longer afford to treat revenue integrity as a problem to be addressed after a claim is denied. As reimbursement pressures intensify and both payers and providers turn to artificial intelligence to analyze claims, identify anomalies and automate processes, health systems need a more proactive approach to protecting revenue while maintaining compliance.
Ritesh Ramesh, CEO of MDaudit, believes that shift requires more than deploying new technology. It means bringing data, people and processes together across coding, compliance, auditing and revenue cycle operations, while using AI where it can deliver measurable business value.
Ramesh recently spoke with Electronic Health Reporter about the changing revenue integrity landscape, the growing “AI versus AI” dynamic between payers and providers, the importance of keeping human judgment in the loop, and what healthcare organizations should consider as they invest in AI-driven revenue cycle strategies.
Healthcare organizations have traditionally focused on managing denials after they occur. Why do you believe that approach is no longer sufficient in today’s reimbursement environment?
The reactive approach to denials management is no longer feasible, as the denial volumes, dollars per denial, and adjudication days per claim have grown incrementally over the years. According to our 2025 Benchmark Report, the average amount per coding-related denial increased 28% from 2023 to 2025. Health systems are under tremendous financial pressure due to a tighter reimbursement and policy environment, and a reactive approach will further stretch their already thin cash flow. Leveraging data, insights, and technology will help them proactively identify denial trends, fix issues before claims are paid, and stay two steps ahead of payers.
You’ve spoken about “Revenue Integrity Redefined.” What does that concept mean in practical terms for health systems, and why is now the right time to rethink traditional revenue integrity strategies?
Revenue integrity is the sustained alignment of three outcomes: reimbursement that reflects the care delivered, compliance that withstands payer and regulatory scrutiny, and operations efficient enough to hold both at enterprise scale. When you pursue any one of the three in isolation, the other two erode. With the advent of AI, now is the right time to transform the people and process dimensions so revenue integrity strategies can be successful. Technology can enable outcomes, but it cannot replace the people and process dimensions in the health system revenue cycle. Investments in technology should be balanced with process reengineering and upskilling people in new technologies.
Many healthcare organizations still treat revenue integrity as a department rather than an enterprise-wide strategy. What mindset shift needs to happen at the executive level?
Enterprises that drive successful revenue integrity strategies leverage a scalable technology platform; stand up a cross-functional program across billing compliance, coding, and revenue cycle; share insights and knowledge; track KPIs that matter; and make measurable progress toward the three outcomes: optimal reimbursement, compliance, and operational efficiency. They don’t see a successful revenue integrity program as a short-term, transactional approach to get claims paid today; they focus on fixing difficult, long-term processes and data issues across the revenue cycle continuum, so these issues don’t recur.
Some cross-functional programs need a strategic charter, actionable KPIs, change management, and business sponsorship to break departmental silos and office politics. This is where executives can drive their revenue integrity vision for the enterprise and align their respective teams to play together as one team.
AI and the Future of Revenue Cycle
Payers are increasingly using artificial intelligence to review claims and identify anomalies. How should providers respond to this changing landscape?
Providers must centralize their claim, payment, and clinical data and glean insights to identify issues and trends before payers report them. A continuous monitoring program and investments in technology platforms that can identify billing, coding, and payment anomalies in real time are the needs of the hour. This will allow them to respond to payers’ use of AI and their investments in this area in a timely and effective manner
AI has become one of the biggest topics in healthcare technology. Where do you see it delivering the greatest value in revenue cycle management today, and where do you think expectations may be getting ahead of reality?
AI has the highest potential to eliminate many manual administrative tasks and automate many parts of the healthcare revenue cycle. Many industry analysts see a world where the cost-to-collect metric will go down for health systems over the next decade because AI and agents help them keep more profits to invest in patient care. Lots of folks hear the word “autonomous” and think AI is a magic wand. I urge them to view AI like any other technology: it needs process and people to succeed. Many AI projects fail because of poor process design and a lack of talent upskilling.
Workforce shortages continue to challenge HIM, coding, and revenue cycle teams. How can AI and intelligent automation help organizations do more with limited resources while maintaining compliance and coding accuracy?
AI can really help with repetitive and mechanical tasks. It can scale exponentially across millions of rows of data and a voluminous number of tasks. When it goes properly, it’s beautiful to watch. When it fails, errors can be amplified, resulting in heavy financial losses. AI governance is critical. With a shortage of expertise in coding and revenue cycle, automation coupled with process and workflow redesign is critical to free those experts to focus on the parts of the process that require human judgment, including validating AI results and handling high-stakes areas.
Payers are increasingly using AI to identify potential claim issues, while providers are also turning to AI to improve audit readiness and revenue integrity. How is this accelerating “AI versus AI” dynamic changing the role of the human auditor, and why is keeping human judgment in the loop becoming more important rather than less?
AI can never be a substitute for human judgment and clinical expertise. It can be trained on large volumes of data sets to find patterns and help identify issues quickly so health systems can act. Providers must invest in AI to level the playing field and accelerate responses to payer queries and denials. Someone recently told me that AI will help them reduce denials. I replied that AI would help them defend the services provided, but it doesn’t expand the insurer’s risk premium pool, which can only pay a finite number of claims. It’s much more important for providers to defend the services they provided in days than for systems to integrate and analyze data for months. They will be left behind.
There has been a lot of discussion about AI adoption in healthcare, but less about proving measurable value. What does “Meaningful AI” mean in practice, and what should healthcare organizations look for to ensure AI investments deliver ROI, reduce operational friction, and strengthen decision-making rather than simply add another technology layer?
Meaningful AI is a framework with five components: data, models, workflow, security controls, and humans in the loop. The whole goal of this framework is to be pragmatic in leveraging AI in use cases with tangible business value. This can be reducing process friction, capturing revenue, or reducing risk. When designing an AI system, people should say, “This use case is not fit for AI because it does not produce any tangible benefits.” Meaningful AI brings that scrutiny and clarity to where AI is applied and what it produces for end users.
Leadership and Industry Outlook
Many healthcare organizations struggle with disconnected data across coding, CDI, compliance, auditing, and revenue cycle teams. How important is breaking down those silos to improving financial performance?
Organizations that achieve strong revenue integrity outcomes do so by combining a scalable technology platform with a coordinated, cross-functional program spanning billing compliance, coding, and revenue cycle operations. They foster knowledge sharing, monitor meaningful KPIs, and drive measurable progress toward three core objectives: optimal reimbursement, regulatory compliance, and operational efficiency.
Rather than treating revenue integrity as a short-term, transactional effort focused solely on getting claims paid, leading organizations take a broader view. They address the underlying process and data issues that create revenue leakage and compliance risks across the revenue cycle, ensuring those challenges are resolved permanently rather than repeatedly managed.
Effective cross-functional initiatives often require a clear strategic charter, actionable KPIs, strong change management, and executive sponsorship to overcome departmental silos and organizational politics. This is where leadership plays a critical role by establishing a unified revenue integrity vision and aligning teams across the enterprise to work collaboratively toward shared goals.
If you were advising a health system CFO making technology investments today, what capabilities would you consider essential for protecting both revenue and compliance?
I would advise CFOs to consider 1) deploying a data driven, proactive strategy to documentation, coding accuracy and denials management; 2) paying attention to payer behavior in real time to understand how best to adapt their RCM strategies; 3) keeping an open mind when reimagining your business processes and upskilling talent while deploying AI; and 4) investing in AI risk and governance as a mandatory function
Looking ahead five years, what do you think revenue integrity will look like, and what changes do you expect will have the biggest impact on healthcare organizations?
As healthcare organizations deploy AI over the next five years, there will be lots of learning from successes and losses. Every health system hopes AI will automate administrative tasks, eliminate waste, and lower cost-to-collect so they can reinvest profits in patient access and care. Various timelines and expectations must be met over the next decade. I am optimistic that there will be many great success stories at the end for others to scale. The current health system infrastructure has been built over decades; it’s unfair to expect AI to perform its magic quickly. I will leave it at that.
Finally, what excites you most about the future of healthcare revenue cycle management and MDaudit’s role in helping organizations navigate that future?
At MDaudit, we are pragmatic about the use of AI in healthcare RCM. We’ve spent a lot of time thinking through which workflows and end users we can impact by leveraging AI. For every idea we accept, we reject 5-10 others.
Our customers work with us as co-creators to develop real products with AI functionality that work in their operational setting. They are the first to tell us what we tried just didn’t work.
We have been embedding Meaningful AI into the MDaudit platform for nearly three years to generate real value. In the last 12 months alone, we’ve generated more than $400 million in value for our customers associated with revenue retention, risk mitigation, and labor productivity.
We see so many future opportunities on our roadmap to continue delivering AI-enabled business value. Our whole organization is both excited about the potential of AI and aware of its risks. Luckily, our customers keep us grounded.
By Comron Saifi, MD, Co-CEO, ClinicFlow. 
