By Rob Ware, senior vice president and general manager of RCM Services, ModMed.
The financial health of medical practices is under strain. Historically, revenue cycle management (RCM) has been treated as a back-office administrative function, but today’s environment of rising operational costs, staffing shortages, and reimbursements that fail to keep pace with inflation demands a new approach.
Adding to the pressure, the industry remains plagued by considerable inefficiency. One study found that insurers, offering qualified health plans through HealthCare.gov, denied 19% of in-network claims and 37% of out-of-network claims. In this situation, administrators spend valuable time going back and forth with insurance companies while patients get hit with surprise bills. Financial stress goes up, institutional trust goes down, and the entire RCM process gets more complicated.
Prevention is the New Strategy
For decades, healthcare organizations have operated under a reactive RCM model. A claim is submitted, a denial occurs weeks later, and staff scramble to identify the problem, correct it, and resubmit the claim. The process has become an expensive cycle of administrative catch-up that drains resources and delays reimbursement.
But healthcare can no longer afford to manage revenue after the fact.
Predictive RCM represents a shift from revenue recovery to revenue prevention. By leveraging historical data and AI-powered insights, practices can identify potential issus before a claim is submitted.
Think of it as a navigation system for the billing workflow. Rather than notifying a user that they missed a left turn after they are already lost, predictive technology highlights potential roadblocks, such as missing authorization requirements, eligibility issues, or coding inconsistencies, while a claim is still being prepared.
The result is fewer preventable denials, faster reimbursement, and less time spent fixing avoidable mistakes.
Empowering the Back Office, Supporting the Front
One of the most important shifts occurring in healthcare today is strengthening the connection between clinical and financial workflows and teams.
Historically, those functions have operated in separate worlds. Yet many reimbursement challenges originate long before a claim reaches the billing department. Missing documentation, incomplete patient information, authorization gaps, and coding discrepancies can often be detected and addressed at the top of the patient journey.
Predictive RCM can help close this gap by identifying claims at risk of denial earlier and creating greater alignment between front-office, clinical, and billing teams. Rather than treating denials as isolated billing problems, providers can address root causes before they impact reimbursement.
This proactive approach not only aims to improve financial performance but also to reduce administrative friction across the practice.
When technology handles highly repetitive and predictable tasks—such as monitoring claim status, identifying missing information, or validating payer requirements—it allows revenue cycle professionals to focus on higher-value work, including complex appeals, payer negotiations, and strategic financial planning.
The benefits extend beyond the back office.
When front-office teams have access to accurate eligibility verification, timely coverage information, and reliable patient cost estimates, they are better equipped to have clear and compassionate financial conversations with patients. Instead of uncertainty, patients receive greater transparency and a clearer understanding of their financial responsibilities before care is delivered.
By shifting from reactive firefighting to predictive prevention, practices can potentially avoid costly denials while giving staff valuable time back.
The Human Impact
And then there’s the patient. Few experiences create more frustration for patients than receiving an unexpected bill months after an appointment.
As patients assume greater responsibility for healthcare costs, financial transparency is becoming an increasingly important part of the overall care experience. Providing accurate information upfront helps reduce anxiety, improve trust, and strengthen the provider-patient relationship.
When patients understand their coverage, expected costs, and payment options before treatment, practices are better positioned to create a more positive financial experience while reducing confusion and collections challenges later.
Ultimately, predictive RCM is about more than preventing denials. It represents a broader shift in how healthcare organizations think about financial operations.
In the years ahead, recovering revenue quickly will be only one piece of the financial success puzzle. Providers will be able to prevent revenue leakage before it occurs, align clinical and financial workflows more effectively, and create a more transparent experience for both staff and patients.
There is a version of a radiology report I have seen hundreds of times. The imaging is done well. The finding is documented. And the report ends with a sentence along the lines of: “Please correlate clinically. Further imaging may be warranted.”
That sentence sounds reasonable. It is also, in many cases, clinically useless. It does not say what should happen next. It does not indicate urgency. It does not tell the ordering provider whether this is a finding that needs action in two weeks or two years. This is called “hedging.” It’s defensive language designed to limit liability without committing to a specific recommendation. And it is one small symptom of a much larger structural problem in imaging.
Radiology-specific AI has made significant gains in detection. Algorithms are getting better at identifying lung nodules, incidental lesions, and abnormalities that might have been missed a decade ago. That progress is real, and it matters. But most of the industry conversation around imaging AI has focused on the front end of the workflow—what the model can find, what it misses, and the usefulness of AI enabled detection—while largely ignoring the back end: what happens after the finding hits the report.
That back end is where care actually breaks down.
The Downstream Problem Nobody Designed For
Every flagged finding is the beginning of a workflow. A follow-up study needs to be ordered. A patient needs to be contacted. A referral may need to be placed. A timeline needs to be tracked. When the finding is serious, those steps carry genuine clinical urgency. When they do not happen, patients get lost.
The data on this is sobering. Research published in the Journal of the American College of Radiology found that overall adherence to recommendations for additional imaging of incidental findings was just 39.1%. Other studies put the figure closer to 50 percent. However you measure it, the gap between what is found and what gets followed up on is enormous, and it widens as imaging volume grows.
And volume is growing. The Neiman Health Policy Institute projects that imaging utilization could increase by as much as 26.9% by 2055, while radiologist supply is expected to grow at a roughly comparable rate, meaning the current shortage is unlikely to improve without deliberate intervention. Radiologist attrition has accelerated since the pandemic, with departure rates up 50% from pre-COVID levels. Under that kind of pressure, report language gets less specific, recommendations get more vague and the downstream infrastructure (which was never adequate to begin with) absorbs more volume than it can handle.
This is the paradox at the center of imaging AI right now. Better detection tools surface more findings. More findings generate more downstream work. And the hard task of translating a finding into actual care relies on a workforce and systems already running at capacity.
More Dashboards Will Not Solve This
Health systems have tried to address the follow-up gap with worklists, tracking spreadsheets, and manual processes. I have watched care navigators spend hours every morning reconciling data from radiology systems against the EHR to figure out which patients still need to be contacted. In many organizations, that manual reconciliation is not a temporary workaround. It is the process. It is also the reason people fall through the cracks.
The limitation of most existing approaches is that they create visibility without creating accountability. A dashboard can tell you that a lung nodule was flagged. It cannot often tell you whether the follow-up was ordered, whether the patient was contacted, whether the appointment was scheduled, or whether the result came back. Those are different operational problems, and each one requires a different handoff.
What is needed is infrastructure that connects detection to completed care. Not just a view of what was found, but an operational layer that routes findings based on actual clinical risk, manages outreach, tracks completion, and escalates when something stalls.
What Completed Care Actually Looks Like
Radiology is often the starting point for a patient’s journey through the health system. From the moment an image is captured to the next step in care, each pathway is different. Patients have different needs, priorities, and resources. Technology workflows have different gaps that require different levels of support to close. Providers serve different populations with different barriers to care. There is no single worklist, workflow, or outreach strategy that can reliably solve every situation, every time.
In an environment defined by high variability and high stakes, high reliability becomes essential. It requires layered processes that apply the right tools to the right problem, with the goal of ensuring that no patient falls through the cracks. This means shifting our focus from task completion to patient outcome. Instead of asking, “Did the provider receive a notification?” we ask, “Did the patient receive the right next step in care?”
That distinction matters. True follow-up requires accounting for the complexity of the patient journey, including the reality that the right next step may change as new information, barriers, or circumstances emerge. High reliability is not measured by whether a task was checked off a list. It is measured by whether the system produced the intended action and result for the patient.
The Real Question for Imaging AI
The radiology AI market has spent the last several years racing to build better detection. That was the right starting point. But the industry is now at a point where the bottleneck is no longer whether a finding can be identified. The bottleneck is whether a finding, once identified, reliably reaches the right clinician, generates the right action, and results in completed care.
Health systems that invested heavily in AI detection tools are beginning to discover that the return on those investments depends almost entirely on what happens after the algorithm runs. A finding that surfaces in a report but never reaches the patient is not a detection success. It is a care failure that started with accurate imaging.
The next chapter of imaging AI needs to be about care completion: building the infrastructure between the radiology report and the EHR, between the finding and the follow-through, between what was identified and what was actually done about it. That is where patient safety lives. And right now, for too many health systems, it is also where patient safety breaks down.
AI is shaping the future of how healthcare organizations manage data, whether they’re ready or not. According to new research, 41% of healthcare organizations are already using AI for database management purposes, with a further 40% considering integrating it soon.
Many practices are already finding value in leveraging AI for their operations, with top applications including data quality assurance, automating database management, and data modeling.
While AI has the ability to generate massive upside for efficiency, it can also wreak havoc across existing data estates if they’re not properly prepared for adoption and integration. When piloting a new AI initiative, it’s imperative that there’s a solid foundation for the model to work on top of. An unstable base could topple down in an instant, unraveling years of work.
Where DBAs should look first
Database administrators (DBAs) must take stock of the key issues with their estate and address them before AI is added into the system. The keys to successful AI adoption can be easily broken down into three key categories: people, process, and data.
DBAs need to first ask if their team is ready to adopt AI. If the humans overseeing it aren’t prepared, then your initiative could fail before takeoff. When timelines are compressed to meet ROI projections set by stakeholders. That means training people with the skills to use AI and the freedom to deploy what they learn in the workflows they are familiar with. Top-down AI usage mandates are not going to help.
Next, DBAs must have a strong grasp on how value flows throughout the organization. Understanding key bottlenecks, which processes are load-bearing, and how to achieve measurable operational outcomes is essential to AI success. Without clarity, AI can be implemented in the wrong places, cascading chaos. It’s easy to point it at a problem that generates no value, or have it contribute to meaningless metrics rather than real outcomes. And fixing the current ones will not be sufficient. Once the first bottleneck is resolved, new ones will emerge that need to be addressed.
Finally, the most critical problem is the data itself. Healthcare databases can be enormous. Estates and their management processes are often passed down from managers from past years or decades. These legacy processes can lead to platforms that are a jumbled mess of software that doesn’t work together, programs that can’t communicate with each other, fragmented estates, undocumented schema changes, and split ownership. AI doesn’t magically clean up these problems, it simply acts as if there’s nothing wrong. If you don’t create a good foundation for AI to operate, then it will churn out confident answers based on broken information, providing solutions that generate no value
Building real foundations
Database governance should be the top priority for any DBA who’s looking to deploy AI. Right now, nearly 40% of all healthcare practices operate across 4 or more database platforms. The best way to address problems of database fragmentation and software sprawl is to pull everything together under a single umbrella, offering a unified view. Without full visibility, issues quickly turn into costly downtime, impacting revenue and customer satisfaction.
Addressing problems with the database’s structure is only half the battle. Once DBAs have cleaned up issues from the past, they must prepare for the future. The most crucial step is to create clear management processes so teams are aligned. Fragmentation occurs when there’s no standardized process for deploying changes or creating pathways. DBAs need to set clear guidelines for deploying updates and tracking schema developments. When engineers have no guiding principles, they create sprawl which could decimate AI processes down the line.
It might be a pain in the short term, but DBAs who dedicate the time to clean their data estate will realize exponential value down the line.
Looking forward
AI is set to revolutionize the way healthcare data is managed. It has the potential to quickly anonymize massive datasets, streamline database management, design schema, and much more.
However, most practices aren’t prepared to realize AI’s true value, and many will suffer due to poor implementation. DBAs need to be cognizant of the foundations that AI needs to thrive, audit their team’s ability to work with AI, identify the bottlenecks within their organization, recognize which internal processes are load bearing, understand how to generate measurable outcomes, and scrutinize the data itself.
AI can only thrive within clear governed processes and solid support. Don’t fall into the trap of thinking it will automatically fix everything.
By Leigh Burchell, vice chair, EHR Association Information Blocking Compliance Task Force.
The HTI-5 proposed rule, Health Data, Technology, and Interoperability: ASTP/ONC Deregulatory Actions To Unleash Prosperity, includes several significant updates to information blocking compliance provisions. The proposed changes raised red flags because they increase compliance challenges rather than providing the simplification, guidance, and education needed to cut through the complexity of current policy.
The EHR Association, which represents nearly 30 health IT developer companies whose technologies support the vast majority of hospitals and ambulatory providers across the US, has several additional overarching concerns, including ASTP/ONC’s overstated predictions about the burden-reduction outcomes of its proposed changes and its underestimation of the true economic impact of both the current and proposed information blocking policy. The reality is that the proposed changes will increase the administrative burden on software developers and other stakeholders who interact with our community as they determine the best path forward for accessing, exchanging, and using information, and the agency should have conducted and included an economic impact analysis of the implementation costs that will be borne by the industry if the proposed changes are finalized.
In the months prior to the proposed rule’s issuance, ONC leadership made it clear that information blocking enforcement was entering a new era. Dr. Thomas Keane, the National Coordinator for Health IT, has repeatedly emphasized since then that certification status and information blocking behavior are linked and that certification nonconformity will be a powerful enforcement tool.
As EHR developers, we support the intent of the information blocking prohibition: seamless information sharing and nationwide interoperability. However, ONC’s policy is not achieving that intention.
Our analysis of HTI-5’s information blocking proposals identified new ambiguities, administrative requirements, and risks for developers and providers. In addition, the HTI-5 proposed rule downplayed the operational and economic impact for covered actors.
Exacerbating the Already-Complex Infeasibility Exception
Among the proposed changes to the Infeasibility Exception are the removal of the “third party seeking modification use” condition and an increase in the number of alternative manners required before an actor can claim the Manner Exception is exhausted.
Most developers rarely use the Infeasibility Exception, yet HTI-5 explicitly implies misuse. Rather than conclude that their policy is overly complex and difficult for even sophisticated actors to understand, ONC’s language in the proposed rule was inflammatory. Real-world examples would be more helpful for the industry to understand how regulators interpret the exception and better allow stakeholders to react accurately and specifically.
The proposal also overlooks the fundamental operational reality that — as the EHR Association has noted since the infeasibility exception was initially proposed — the 10-business-day response window is simply unworkable. Understanding and scoping a connectivity request, reacting to the original manner requested, and then possibly assessing and negotiating up to three alternative manners cannot be completed within that timeframe. For many health IT developers, especially smaller vendors, the volume of requests alone makes this impossible. We have long recommended amending the regulation to require that the 10-day clock begin after negotiations conclude, not upon receipt of the request.
A Shift Toward Ambiguity in Manner Exception Exhausted
HTI-5 would replace Manner Exception Exhausted wording that actors offer the “same” access, exchange, or use with an “analogous” one. While seemingly minor, this change introduces significant ambiguity. What is “analogous” in one system architecture may not be in another. ONC’s own example implies that all write APIs or filters are inherently analogous, which notably oversimplifies the diversity of technical implementations across the industry.
This ambiguity complicates both compliance and enforcement. OIG and private litigants increasingly cite information blocking exceptions in disputes. Subjective terminology raises the stakes for everyone, particularly in private litigation, where judges frequently lack the requisite technical knowledge to assess the situation accurately. That is why we recommend retaining the current “same” standard.
Putting Innovation at Risk
HTI-5 also proposes that if any entity receives a requested manner of access, exchange, or use, the Manner Exception Exhausted cannot apply. This would discourage technical pilot projects, custom development for healthcare organizations, and early-stage innovation — precisely the activities that advance interoperability. If a single-pilot implementation sets a precedent for all future requesters, developers could be far less willing to experiment.
We strongly recommend to ONC that the Manner Exception Exhausted condition be retained and improved by both clarifying expectations for negotiation and setting more realistic timelines.
New Manner Exception Definitions, Burdens, and Conflicts
HTI-5 attempts to clarify the Manner Exception by introducing new requirements around market rate, contracts of adhesion, and general market value. Unfortunately, these changes create more confusion than clarity and risk slowing processes specific to innovation and contracting.
Fair Market Value (FMV) requirements are u Defining “market rate” as Stark Law FMV would require developers to obtain formal valuations for potentially every product, module, or custom integration. ONC did not estimate the costs of this often-lengthy process, but they would be substantial and would slow innovation, including emerging AI-driven capabilities.
Contract-of-Adhesion language conflicts with other ONC r ONC requires standardized, publicly posted API terms and pricing. Yet HTI-5 suggests that standardized terms may be considered contracts of adhesion. Developers cannot be simultaneously required to publish consistent terms and penalized for using them.
Flexibility for custom work is eliminated. Removing paragraph (a)(2) from the Manner Requested Condition and forcing all arrangements into the Fees and Licensing Exceptions would undermine the fundamental purpose of the Manner Exception. Custom integrations often require custom pricing. Restricting that flexibility would discourage innovation, limit vendors’ flexibility to meet clients’ unique requests, and reduce the availability of unique, mutually beneficial exchange arrangements.
Burden Estimates: A Missing Piece of the Regulatory Puzzle
Our assessment is that the information blocking elements of HTI-5 are, in fact, the opposite of deregulatory, given the burden they would impose. Despite introducing the possibility of new negotiation expectations, valuation requirements, and interpretive standards, ONC provided no economic impact analysis for these proposals in the HTI-5 NPRM. The absence of these estimates is especially concerning, given the disproportionate impact on small developers and the cascading compliance obligations for both providers and third-party partners.
Accurate burden estimates must include:
Expected volume of Manner Exception negotiations
Time and staffing required for multi-step negotiations
Costs of FMV valuations
Impact on innovation, especially AI and custom integrations
Without this analysis, HTI-5 proposals cannot reasonably be considered deregulatory. We suggest that the proposals specific to information blocking not be finalized without additional rulemaking proposals that include economic impact assessments, provide an opportunity for industry input, and verify the accuracy of that analysis.
The EHR Association suggests that the information blocking sections of the HTI-5 NPRM require significant rework to achieve greater clarity, practicality, and accuracy before any related concepts are finalized. We strongly support a regulatory framework that encourages, rather than constrains, innovation and interoperability, but the current proposals are not yet there.
When a hospital’s systems go dark, the danger doesn’t stay in the server room. It moves to the bedside.
That’s not a hypothetical. Recent threat intelligence found that healthcare organizations experienced a cyberattack roughly every 10 hours between January 2025 and February 2026 — the highest incident rate of any sector analyzed. Ransomware alone accounted for nearly 60% of those attacks.
HBO’s “The Pitt” dramatizes exactly what that looks like in practice. When two nearby hospitals are hit by a cyberattack, the fictional Pittsburgh Trauma Medical Center shuts down its connected systems to contain the threat. The digital patient board goes dark. Doctors revert to paper charts. Medication orders are delayed, lab results go missing, and clinicians are left making time-sensitive decisions without the patient histories they depend on. A missed life-threatening diagnosis follows.
The show is fiction. The operational risk it depicts is not.
Downtime Is a Patient Safety Problem
Healthcare has become an attractive target because disruption creates immediate pressure. Attackers understand that hospitals depend on continuous access to data, systems and connected devices. They also understand that downtime can affect patient flow, procedures, pharmacy operations, lab ordering and clinical decision-making.
The healthcare threat intelligence report describes healthcare as a sector with “life-or-death operational dependency,” high-value protected health information, chronic security underinvestment and complex legacy infrastructure. That combination makes hospitals vulnerable to attacks that affect both data security and care delivery.
When systems go down, the effects ripple across the organization. Ambulances may be diverted, procedures may be delayed or canceled, pharmacy systems may become unavailable and clinicians may lose access to electronic health records, prior diagnoses, medication histories, allergies and test results.
In a hospital, those are the foundations of safe, coordinated care. Cyber threats, therefore, carry greater risk than routine workflow interruptions.
“The Pitt” illustrates this dynamic by focusing on the mechanics of downtime. The tension comes from clinicians trying to work without the information and processes they normally rely on. Paper charts replace digital records. Verbal handoffs replace system visibility. Manual steps replace automated safeguards.
This is where healthcare leaders can focus their efforts. One takeaway from the show is not that hospitals should fear a dramatic ransomware scenario. The lesson is that downtime readiness must be treated as part of patient safety planning.
The Weak Points Are Often Familiar
Attackers don’t need sophistication, they need an opening. In healthcare, those openings are rarely exotic. The most common entry point is authentication bypass: flaws that let attackers reach privileged systems without proper credentials. In an environment where dozens of platforms, vendors, contractors and devices all need access to keep care moving, that risk compounds quickly.
The pattern that follows is predictable. A weakness in one layer – an unpatched remote access portal, an overlooked vendor credential, a known vulnerability that never got remediated – creates a failure somewhere else entirely. Lab ordering goes down. Pharmacy systems become unavailable. Imaging access disappears. What began as a security incident becomes a clinical one.
Every tracked vulnerability in our analysis appeared in the CISA Known Exploited Vulnerabilities catalog. Securin’s latest healthcare threat report makes the implication hard to ignore: the sector is overwhelmingly exposed to vulnerabilities we already know how to fix. That’s not a resource problem, it’s a prioritization one. Attackers follow the path of least resistance, and known, unpatched vulnerabilities remain valuable precisely because they persist in operational environments long after they’re publicly disclosed.
The report also found that many healthcare organizations, under pressure to restore operations quickly, continue to pay ransoms. That calculus is understandable at the moment, but it funds the next attack. Healthcare’s combination of operational urgency and chronic security underinvestment has made it the most reliably profitable sector for ransomware operators.
Cyber Resilience Has to Include Clinical Downtime
Preventing intrusions matters, but it’s not enough. The harder question for healthcare leaders is this: when critical systems become unavailable, can your hospital keep delivering care safely?
That question exposes a gap in how most organizations think about cyber risk. Security controls live in the IT department. Downtime procedures, if they exist, often live in a binder somewhere. But the consequences of a cyberattack play out in the ED, the pharmacy, the lab and the OR. Resilience planning has to reflect that.
The vulnerabilities most likely to cause hospital-wide disruption are well known: internet-facing systems, remote access tools, identity and authentication platforms, and administrative interfaces. Addressing those isn’t glamorous work, but leaving them unpatched while investing in more sophisticated defenses is like reinforcing the roof while leaving the front door open.
Operationally, the gap between security and care delivery has to close. Downtime procedures should be practiced with the people who actually deliver care – clinicians, nurses, pharmacists, lab teams – not just tested in an IT tabletop exercise. Teams need to know how to place paper orders, reconcile medications, track patients and hand off information safely when digital systems aren’t available. When systems come back online, the process of restoring and reconciling that information carries its own risks.
The Bedside Is Now Part of the Cyber Risk Model
The most frightening moments in “The Pitt” are not the attack itself. They are the human ones that follow: a missing patient history, a delayed medication order, a clinician making a life-or-death decision with incomplete information. The show resonates because it understands something that healthcare security teams have been trying to communicate for years – that in a hospital, a cyber incident is never just an IT problem.
Healthcare leaders cannot assume every attack will be prevented. The threat intelligence is too consistent, the attack surface too broad and the incentives for attackers too strong. But prevention is only half the mandate. The other half is ensuring that when systems fail -and some will – care teams can keep patients safe anyway.
That requires security fundamentals: closing the known vulnerabilities attackers are already exploiting, enforcing stronger access controls, segmenting networks so one compromised system doesn’t become a hospital-wide crisis. It also requires something harder to operationalize – a genuine integration of cyber resilience into patient safety planning, tested with the people who deliver care, not just the people who manage infrastructure.
When connected systems go dark in a hospital, the consequences move fast. A missed diagnosis. A lost order. A bad handoff. The gap between a cyber incident and a patient safety event can close in minutes.
By Dr. Wael Khouli, MD, MBA | Co-Founder & CMO, Authsnap, Inc.
I just came off the stage at NPAC 2026 in Charlotte, where I presented to a room full of physician advisors on practical AI strategies for expanding bandwidth. The questions afterward told me everything I needed to know about where this profession stands right now.
Nobody asked whether AI was coming. They already know it is. They asked something more honest than that: how do we use it without losing the thing that makes us valuable in the first place?
That is the right question, and it deserves a real answer.
The Volume Problem Nobody Wants to Say Out Loud
Physician advisors are being asked to do more reviewing, more documentation, more peer-to-peer justification, and more appeals work than at any point in this profession’s history. Payer policies are more granular. Medicare Advantage utilization management has expanded into services that were once routinely approved. The documentation bar for medical necessity keeps rising.
The volume that now lands on a physician advisor’s desk has outpaced what any human can sustainably handle at the quality this work demands. That is not a criticism. It is arithmetic. And it is the honest starting point for any serious conversation about AI in this space.
The physician advisors I talk to are not afraid of AI replacing them. They are exhausted by a volume of work that was never supposed to be theirs, and they want to know whether AI can take some of it back.
There is a second category that gets even less attention: the work that never gets done at all. The cases that go unreviewed, the appeals that are never filed, the denials written off because there was no bandwidth to fight them. That work does not show up in a productivity report, but it is where revenue and patient access quietly leak out. With AI assistance, physician advisors and utilization management specialists can finally reach the work they could never get to before, not just move faster through the work already on the desk.
What AI Can Actually Do, and What It Cannot
Let me be specific, because vague claims about AI in healthcare help no one.
AI is genuinely good at the extractive, pattern-matching layer of physician advisor work: ingesting a clinical record and identifying the relevant diagnoses, treatment history, and documentation gaps; cross-referencing that information against payer-specific criteria; and generating a structured, evidence-based argument that a physician advisor can then review, refine, and sign off on. Done manually, that work takes one to two hours per case. AI brings it under ten minutes.
What AI cannot do is the clinical judgment underneath. Reading what a patient presentation actually means. Knowing when a payer’s stated criteria do not reflect the clinical reality of a complex case. Understanding the institutional context that shapes a particular denial pattern. Recognizing when the documentation tells a different story than the billing codes. That layer is not automatable, and any tool that claims otherwise is either wrong or selling something.
The physician advisors who will define this profession are the ones who understand exactly where that line sits. They use AI aggressively on one side of it and protect their judgment fiercely on the other.
The HIPAA Conversation Nobody Should Skip
I made this explicit at NPAC and I will make it explicit here: responsible AI adoption in physician advisory work requires HIPAA-compliant infrastructure, full stop. Not as an afterthought. Not as a future-state aspiration. Clinical documentation is protected health information, and any AI tool processing it needs to operate in a closed, compliant environment with audit trails and explainability built into the architecture.
The tools that earn physician advisor trust are the ones that make compliance visible, not the ones that ask you to trust a black box. When an organization deploys generic AI that lacks clinical specificity or payer fluency, the result is predictable: appeal quality drops, risk exposure rises, and the physician advisor ends up doing more cleanup than the manual process would have required. That is not a technology failure. It is a selection failure.
What Mastery Looks Like From Here
The physician advisors getting this right share a posture. They treat AI as a tool that earns trust through performance, not as a default answer to a capacity problem.
They verify AI-generated content before it goes out. They understand the clinical logic behind the output well enough to catch it when it is wrong. And they spend the time AI gives back on the work that requires judgment: the peer-to-peer conversations, the complex case reviews, the pattern recognition across a portfolio of denials that signals something systematic happening upstream.
AI bandwidth paired with physician judgment is the standard this profession is moving toward. The advisors who master it will handle case volumes that would have broken the old model. The ones who do not will fall further behind with every payer policy update.
The Honest Bottom Line
AI is not going to replace physician advisors. The clinical judgment, payer fluency, and institutional knowledge that make a great physician advisor valuable cannot be replicated by any model available today. But the administrative layer sitting on top of that judgment, the documentation extraction, the criteria cross-referencing, the initial appeal construction, can and should be automated.
That is not a threat to the profession. It is what gives the profession its future back. The physician advisors who embrace it will do the most important work of their careers. The ones who resist it will spend those same years buried in paperwork.
One caution belongs at the end of this. Moving forward, it will be difficult for physician advisors to stay at the forefront without becoming genuinely proficient at deploying AI in the right way and the right place. That proficiency runs in both directions. It means applying these tools where they create real leverage, and it means refusing to over-rely on them. The advisors who stop validating the output, who let AI quietly absorb the clinical judgment that only they can own, will not lead this profession. They will be exposed by it. Mastery is using AI aggressively while keeping a firm hand on the parts that must stay human.
By Ryan Dewey Smith, founding executive chairman & CEO, Inperium.
Behavioral health and human-centric services delivery is rapidly shifting to digital-first, A.I.-supported, and algorithmic modeling. Far beyond the realm of business functions like automating appointment reminders or facilitating insurance claim submissions, these technologies now are applied in clinical use to augment care.
A.I. holds promise for early medical intervention through its power to analyze full medical records and identify potential causation for patterns it recognizes, and clinicians in behavioral health services are increasingly utilizing digital therapeutic software to deliver evidence-based interventions to diagnose, treat, or prevent mental and behavioral health disorders.
From using A.I. language interpretation programs for clinicians to speak with diverse language populations or applying voice-to-text to generate timesaving summaries of patient sessions to employing A.I.-enabled wearables, there are numerous applications that hold real promise for better patient outcomes.
Like in every sector, emergent technologies are moving into widespread use at lightning pace, and like all emergent technologies, they are not cheap. A.I.-enabled devices share something with analog mobility and communication or sensory tools:
They are complex, difficult to maintain, and expensive. State-of-the-art technologies, whether used in the clinician’s back office or employed as an enhancement to medical care, follow the familiar pattern of being available only to those who have the financial wherewithal to afford them.
The harsh reality is that a large segment of those seeking professional assistance to manage mental health needs, addiction recovery, or disabilities services would never receive care if it were not for nonprofit providers. Our most vulnerable populations are overwhelmingly served by organizations already operating on razor-thin budgets.
Such providers lack the resources to access new technologies. In rural locations many providers have limited access to reliable broadband, an underlying requirement for implementing digital services. This fundamental digital divide is something they have in common with their impoverished patients, no matter their geographical location.
Yet a bridge across this divide could mean so much. Applications like an eye-tracking speech and visual output device or a smart medication dispenser that can track doses and send reminders are life-changing. We should be excited for the transformative innovations that are being developed, but the current structure threatens to worsen the existing disparity of access to them. Why should those with the greatest needs get left behind?
As the larger culture is carried on a tsunami into the A.I. future, we are reminded that we will require an equally sized tidal wave of education to use it effectively. As expensive as technology can be to purchase or use, that price pales compared to the human capital required to learn to use it well.
Successful adaptation will require an advanced set of critical thinking skills, immersive training in best A.I. practices, and heightened attention on data security measures to protect patient privacy. Not only will there be a learning curve, but there will also be vital data transition required, for in underfunded nonprofits, much of the treasure trove of data is stuck in analog systems (including paper record systems) and digital systems are often either outdated or cannot speak with one another. A.I. is useless with incomplete, inaccurate, or obsolete data.
The need for provider education is only half the equation. Reliance on patient usage of systems requires access to the technology and the know-how to use it. Many populations served lack one or both. Should we make the considerable investment required to realize all the benefits that these technologies can bring to the sector, it won’t have any meaning if we are not successful in providing the education our patients will require.
As with most difficult human problems, education can’t stop at “this is how it works;” it has to be adopted, which means investment in the time required to assist patients in altering their lifestyles.
People naturally fear the presence of “robots” in the parts of their lives that make them most vulnerable. Let’s learn the lessons of the COVID Pandemic, that a lack of human connection creates vast unintended consequences. Humancentric services can lead this transition by maintaining human connection while helping those they serve capture the benefits that nonhuman assets can provide.
Only human professionals can ensure that the most vulnerable among us are not made more vulnerable as we move into a technologically enhanced future.
Mealtime insulin is a routine part of diabetes care, but it also demands careful coordination among patients, prescribers, pharmacists, and caregivers. In that wider system, CanadianInsulin is a prescription referral platform; where required, it helps confirm prescription details with the prescriber, while dispensing and fulfilment are handled by licensed third-party pharmacies where permitted.
Some patients explore cash-pay options and cross-border fulfilment depending on eligibility and jurisdiction. Those system issues should remain separate from clinical decisions, which belong with a qualified healthcare professional who knows the patient’s diagnosis, glucose patterns, and treatment history.
Why Mealtime Insulin Pens Matter In Diabetes Care
Insulin pens are designed to help deliver measured doses in daily life. For many people, they are easier to carry and use than a vial and syringe. They may also support more consistent routines when meals, work, school, or travel make timing difficult.
Humalog KwikPen is one example of a prefilled insulin pen used for mealtime insulin. The important clinical point is not the pen alone. It is the combination of insulin type, dose instructions, meal timing, glucose monitoring, and follow-up care.
Rapid-acting insulin can lower blood glucose quickly. That makes it useful around meals, but it also means mistakes can have fast consequences. A missed meal, extra dose, unusual activity, or illness can change the risk of low blood sugar.
What Rapid-acting Insulin Lispro Does
Humalog is a brand name for insulin lispro. Insulin lispro is a rapid-acting insulin analog. It is made to start working faster than regular human insulin, so it is often used near meals or to correct high blood glucose when prescribed.
So, is Humalog KwikPen the same as insulin? More precisely, it is a disposable prefilled pen that contains insulin lispro. The medication is insulin lispro; the KwikPen is a delivery device. This distinction matters because patients may receive the same insulin in different formats, such as pens, cartridges, or vials.
What kind of insulin is HUMALOG 100? HUMALOG U-100 contains 100 units of insulin lispro per milliliter. U-100 is a concentration, not a dose. A prescribed dose still depends on the individual plan written by the clinician.
Rapid-acting insulin is commonly used with a longer-acting insulin in people who need basal and mealtime coverage. Some people with type 2 diabetes use mealtime insulin only after other treatments no longer provide enough control. Others, including many people with type 1 diabetes, require insulin as a central part of daily survival.
Where Pens Fit In tTreatment Decisions
A pen can simplify the mechanics of injection, but it does not decide the treatment plan. Clinicians consider age, diagnosis, kidney and liver function, eating patterns, glucose data, other medicines, vision, dexterity, and support at home.
Some patients need fixed mealtime doses. Others use insulin-to-carbohydrate ratios or correction scales. These approaches require education and periodic review. A plan that worked months ago may not fit after weight changes, new medicines, pregnancy, illness, or changes in activity.
Pens can be helpful for people who need discreet dosing outside the home. They may also reduce some measuring steps compared with syringes. Still, pen use requires training. Patients must understand how to attach needles, prime the pen if instructed, select the prescribed dose, inject correctly, and dispose of needles safely.
Device choice may also be affected by coverage rules, local availability, prescriber preference, and pharmacy substitution policies. These are administrative factors, but they can affect continuity of care. Patients should tell their care team before switching formats or concentrations.
Safety Checks that Reduce Avoidable Errors
The most important safety risk with rapid-acting insulin is hypoglycemia, or low blood sugar. Symptoms may include shakiness, sweating, fast heartbeat, confusion, hunger, weakness, or irritability. Severe hypoglycemia can cause seizures, loss of consciousness, or injury.
Patients should ask their clinician how to prevent, recognize, and treat low blood sugar. They should also know when to use emergency glucagon if it has been prescribed. Driving, exercise, alcohol, delayed meals, and overnight routines may need specific planning.
Dosing errors are another major concern. Insulin products can have similar names or packaging. Concentrations can differ. A U-100 product should not be confused with more concentrated insulin unless a clinician has clearly changed the prescription and provided education.
Several basic checks can reduce risk:
Confirm the insulin name and concentration before each new supply is used.
Use the pen only for the person it was prescribed for.
Do not share pens, even if a new needle is attached.
Rotate injection sites to reduce lumps or skin changes.
Use a new needle as instructed and dispose of it in a sharps container.
Follow storage instructions and ask a pharmacist about heat, freezing, or travel concerns.
Patients should report repeated lows, unexplained highs, injection-site problems, allergic symptoms, or confusion about technique. They should not adjust mealtime insulin on their own unless their care plan clearly explains how to do so.
Access Questions Belong In the Care Pathway
Access to insulin is not only a pharmacy issue. It is also a continuity-of-care issue. Delays, coverage changes, prior authorizations, travel, or prescription mismatches can interrupt treatment and raise clinical risk.
Patients can reduce disruption by keeping an updated medication list and knowing the exact insulin name, concentration, device type, and prescriber instructions. Caregivers should know where supplies are kept and what to do if doses are missed or blood glucose becomes unsafe.
When administrative questions arise, the prescriber’s office and pharmacy are usually the key sources of clarification. They can confirm whether a substitute is clinically appropriate, whether a prescription needs revision, and whether device training is required.
Patients should also ask how much insulin to keep on hand within the limits of their prescription and local rules. Planning is especially important before holidays, severe weather, travel, or changes in insurance status. The goal is not stockpiling; it is avoiding preventable gaps in a medication that may be time-sensitive.
Summary for Patients and Caregivers
Mealtime insulin pens sit at the intersection of clinical care and everyday logistics. A pen can make dosing more portable, but safe use depends on the correct insulin, clear instructions, glucose monitoring, and timely communication with the care team.
Humalog KwikPen contains insulin lispro, a rapid-acting insulin used around meals or as otherwise prescribed. HUMALOG U-100 refers to a 100 units per milliliter concentration. Patients should not treat concentration, device format, or brand name as interchangeable without professional guidance
This content is for informational purposes only and is not a substitute for professional medical advice. Anyone using insulin should follow the plan provided by their healthcare professional and seek urgent care for severe low blood sugar, serious allergic symptoms, or unsafe glucose readings.
