How one baby’s extraordinary medical journey challenges the way we think about physician burnout, medical billing, administrative inefficiency, and artificial intelligence.

“Recruitment brings doctors to North Bay. Time with patients, not paperwork, keeps them here.” — Thom Brott, Chair of the Board of Directors of MetricAid, in a letter published by BayToday on October 7, 2026.

Healthcare keeps inventing tools to make work faster. But what if we’re automating the wrong problems? Wren’s story invites us to question a system that measures activity obsessively while sometimes losing sight of the people it exists to serve.

A Baby Who Challenged the Prognosis

In Montz, Louisiana, Nick Roberts and Savannah Roberts faced a possibility no parent wants to imagine: their daughter might not survive birth.

Their baby, Wren Michelle Roberts, had been diagnosed with an extraordinary combination of congenital conditions, including heterotaxy syndrome, spina bifida, and multiple heart defects.

The prognosis was devastating. Her parents were warned that she might not survive birth. They even began preparing for her funeral.

Then Wren was born in September 2026.

She survived. She underwent spinal surgery at one week old. She required a feeding tube and continued to face serious medical challenges. Yet she also began reaching developmental milestones her family had feared she might never experience.

Her physician, Dr. Gabriella Bluett-Mills, a pediatric complex-care specialist at Ochsner Children’s Hospital in New Orleans, helped guide her care.

As reported by The Guardian on October 8, 2026, Wren’s journey has continued to surprise her family and medical team. Her parents have cautiously introduced her to ordinary experiences outside the hospital, including church and a high school football game.

For another family, these might be unremarkable outings.

For Wren’s family, they carry a different meaning.

Her story is not proof that medical predictions are useless. It is not a promise that every child with a serious condition will defy expectations.

It offers a more careful lesson: a prediction can inform a decision without becoming a substitute for continued observation.

And that principle raises a question for the rest of healthcare.

What happens when the systems designed to support medicine become so focused on their own processes and measurements that they lose sight of what those processes are supposed to accomplish?

Read the original report in The Guardian.

A Quote From This Week That Deserves More Attention

On October 7, 2026, Thom Brott, chair of the board of directors of physician scheduling and coordination company MetricAid, published a letter in BayToday about physician recruitment and retention in Northern Ontario.

He wrote:

“Recruitment brings doctors to North Bay. Time with patients, not paperwork, keeps them here.”

The line is concise. Its implications are not.

Healthcare organizations invest considerable effort in recruiting physicians, expanding training, and addressing workforce shortages. But bringing a physician into the system is only part of the challenge.

The working environment matters, too.

Brott cited the Canadian Medical Association’s 2025 National Physician Health Survey, which found that physicians reported spending an average of 10.4 hours per week on administrative tasks. His letter also noted that 77% of surveyed physicians identified reducing administrative burden as a way to improve recruitment and retention. These are Canadian survey findings, not estimates of U.S. physician workload. Source: BayToday, October 7, 2026.

This is the uncomfortable contradiction.

We spend years training physicians to make complex clinical decisions. We recruit them to care for patients. Then we surround them with systems that can consume enormous amounts of their attention before, during, and after the encounter.

We have become very good at getting people into healthcare.

Are we equally good at making it possible for them to do the work they came to do?

And here is the contrarian question:

What if part of the workforce problem is not simply that healthcare needs more physicians, but that healthcare needs to stop wasting so much of the time it already has?

Recruitment matters. Staffing matters. Compensation matters. Administrative reform is not a substitute for any of them.

But a system that continually loses capacity to avoidable work should not assume that hiring more people is always the first or only answer.

Sometimes, the problem is not a shortage of hands.

It is the amount of unnecessary work those hands are expected to perform.

Healthcare Has a Strange Definition of Progress

Healthcare loves progress.

We measure productivity, throughput, utilization, turnaround times, reimbursement, denial rates, patient volumes, and countless other indicators.

We introduce new software. We automate tasks. We build dashboards. We deploy artificial intelligence that can summarize information in seconds.

Yet in many practices, physicians still finish their clinical work only to begin another shift of documentation, authorization follow-ups, claim corrections, and administrative problem-solving.

We have become remarkably good at measuring activity.

But activity is not the same as completion.

A claim can be submitted without being paid.

An authorization request can be sent without being approved.

A task can be marked complete while the underlying problem remains unresolved.

A report can be generated without changing a single decision.

Somewhere along the way, we began treating the completion of a process as though it were the achievement of its purpose.

That is an expensive mistake.

Imagine a clinic that reduces claim-submission time by 30% but sees no improvement in payment speed, denial rates, or staff workload.

Has the clinic become more efficient?

One step has become faster. That is progress of a kind.

But if the same problems continue downstream, the overall system may not have improved very much.

We may simply have become better at sending problems to the next department.

This is the difference between optimizing activity and improving outcomes.

The first is easier to demonstrate in a software presentation. The second is harder to achieve—and much more valuable.

The Contrarian View: Healthcare May Need Fewer Tasks, Not Just Faster Tasks

Here is an idea that may sound strange coming from the founder of an AI-powered medical billing company:

Healthcare does not necessarily need more automation. It needs fewer avoidable tasks to automate.

This is not an argument against technology. It is an argument for choosing the right problems.

Consider three familiar situations.

A claim repeatedly fails because required information is missing. An AI tool helps staff identify the omission.

Useful? Absolutely.

But what if the information could have been captured correctly during the original encounter?

A billing team spends hours correcting inconsistent patient information. A better correction workflow reduces the time needed to fix those records.

Useful? Certainly.

But what if the inconsistency could have been prevented at the point of entry?

Staff members repeatedly chase incomplete handoffs. A task-management tool makes outstanding work visible.

Useful? Yes.

But what if the handoff could have been completed correctly in the first place?

In each example, the technology may be doing a good job.

The more important question is whether we are asking it to solve the right problem.

Think of a leaking pipe.

One approach is to hire someone to mop the floor faster, another person to monitor the water level, and a third to generate a report whenever the floor gets wet.

All three services might improve.

The floor would still be wet.

Eventually, someone should ask whether repairing the pipe would be a better investment.

Healthcare has its own version of the wet floor: missing documentation, incomplete patient information, unclear ownership, inconsistent workflows, and preventable billing errors.

These problems generate downstream work. Downstream work creates more handoffs. Handoffs create more opportunities for information to be lost or misunderstood.

Then organizations purchase additional tools to manage the consequences.

Sometimes those tools are necessary. But the underlying process deserves scrutiny before another layer of software is added.

The most efficient workflow may be the one that never needs to begin.

The Hidden Cost of Administrative Complexity

For an independent physician practice, administrative friction is not an abstract management problem.

It affects cash flow, staffing, patient access, and the ability to operate sustainably.

A denied claim is not merely a line on a dashboard. Someone may need to investigate the reason, retrieve documentation, correct an error, communicate with a payer, resubmit the claim, and track the eventual payment.

That work consumes time whether or not it produces additional revenue.

The cost becomes difficult to see when the work is distributed across multiple employees, systems, and organizations.

Imagine a small clinic:

• A front-desk employee enters incomplete insurance information.
• A medical assistant notices a discrepancy but does not know who owns its resolution.
• The billing team discovers the problem after the encounter.
• A claim is delayed or rejected.
• Staff members exchange messages to reconstruct the original information.
• The physician is asked to clarify documentation days after the visit.
• Someone follows up again because the first correction did not resolve the issue.

Each person may have performed their assigned task correctly.

Yet the overall process has failed.

This is an important distinction: a system can contain competent people and still produce poor outcomes when its design creates unnecessary friction.

The natural response is often to ask employees to be more careful.

Sometimes additional training is appropriate. But when the same errors recur across multiple employees, the problem may not be individual carelessness.

It may be a workflow that allows incomplete information to move forward, unclear responsibility for resolving discrepancies, or disconnected systems that make essential information difficult to find.

Telling staff to pay more attention is not a complete operational strategy.

Neither is buying software and hoping it will fix everything.

The better question is:

Where did the problem begin, and why was the system able to pass it downstream?

Physician Burnout: What If the Problem Is Not Just Working Too Much?

Physicians expect demanding work.

They train for years to make difficult decisions under pressure. They accept responsibility for uncertainty, complexity, and outcomes that cannot always be controlled.

That does not mean they should have to accept every administrative burden as an unavoidable feature of medicine.

Administrative burden is not the sole cause of physician burnout. Workload, staffing, organizational culture, compensation, clinical demands, and many other factors matter.

But administrative work is an area worth examining because practices can often identify and improve specific operational problems.

There is a difference between a difficult clinical decision and an administrative task that repeatedly demands attention because the system failed to resolve something earlier.

The first may be inseparable from medicine.

The second deserves investigation.

A physician might spend several minutes resolving a documentation discrepancy, revisit the same issue when a claim is denied, and answer another message when the correction is incomplete.

Each task appears small.

The accumulated interruption is not.

And there is a psychological cost to work that never seems to finish. A physician can complete a full day of patient care and still feel behind because the administrative queue has regenerated itself.

The clinic becomes a place where the workday does not end. It merely changes tabs.

We should not assume that every administrative task can be eliminated. Healthcare requires documentation, compliance, coordination, and accountability.

But necessary administration and avoidable rework are not the same thing.

The goal should not be to make physicians better at tolerating broken workflows.

It should be to give them fewer broken workflows to tolerate.

AI Is Becoming More Common. Is It Solving the Right Problems?

Artificial intelligence offers genuine opportunities in healthcare administration.

The Canadian Medical Association’s 2025 National Physician Health Survey, cited in Brott’s October 7 letter, found that 59% of surveyed physicians said AI had already reduced their administrative time. That is an encouraging signal, although it does not mean every AI implementation produces the same results. Source: BayToday.

But adoption is not the same as impact.

A clinic can introduce AI and still have unclear workflows. It can generate better notes and still struggle with incomplete registration data. It can automate a billing task and still lack visibility into why claims repeatedly fail.

The question should not be whether a tool uses AI.

The question should be whether the tool changes the underlying work for the better.

There are three useful levels of AI-enabled improvement.

Level 1: Automate the existing workflow

Technology performs familiar tasks more quickly.

Examples include sorting claims, identifying possible coding issues, drafting correspondence, or organizing follow-up work.

This can be valuable, particularly when the task is repetitive and the process is well defined.

But it does not necessarily prevent the problem from recurring.

Level 2: Detect problems earlier

Technology identifies missing or inconsistent information before it creates downstream rework.

For example, a system might flag a missing required field before a claim is submitted or identify a discrepancy that needs human review.

Earlier detection can reduce avoidable delays, provided the system uses reliable information and its alerts are actionable.

Level 3: Prevent problems at the source

This is the more ambitious approach.

Instead of waiting for the billing team to discover an issue, the practice examines how the issue originates.

Can information be captured more consistently? Can responsibilities be made clearer? Can required details be collected at the appropriate point in the workflow? Can staff avoid entering the same information into multiple systems?

Prevention is not always possible. Payer rules change, information may be unavailable, and some situations require human judgment.

Nevertheless, recurring failures deserve investigation.

The progression should be from processing problems, to detecting problems, to preventing avoidable problems.

The best solution may combine all three.

The Data Quality Problem Hiding Inside Medical Billing

At OnnX, I think about medical billing through a particular lens:

Healthcare billing is not just a software problem. It is also a data-quality problem.

Billing depends on information gathered across a series of interactions: patient registration, insurance verification, scheduling, clinical documentation, coding, authorization, claim submission, and payment posting.

When information is incomplete, inconsistent, or disconnected, downstream teams must compensate.

They search for missing details. They clarify discrepancies. They correct records. They repeat work that should not have been necessary.

Technology can help manage that complexity. But technology cannot reliably transform every incomplete input into a correct outcome.

Garbage in, beautifully summarized, is still garbage.

A sophisticated AI model cannot safely infer every missing fact. A billing system cannot resolve every ambiguity without appropriate information or human review. A dashboard cannot compensate for a process in which nobody knows who is responsible for the next step.

This is why the point at which information is captured matters.

Imagine two clinics with the same billing software.

Clinic A captures required information inconsistently. Staff members use workarounds, handoffs are informal, and discrepancies are often discovered after submission.

Clinic B has clearer responsibilities, more consistent data capture, and a process for identifying exceptions before they propagate.

The software may be identical.

The operational results may not be.

The difference may be the quality of the information entering the process and the reliability of the process that handles it.

For small and medium-sized physician-owned practices, this distinction is especially important. They often have limited staff and less capacity to absorb repetitive administrative work.

They need systems that fit their realities—not another complicated layer requiring constant supervision.

The Most Expensive Word in Healthcare May Be “Again”

Consider how often healthcare teams repeat work.

Verify the same information again.

Request the same document again.

Correct the same field again.

Explain the same denial again.

Follow up on the same unresolved issue again.

Re-enter the same data again.

Each repetition seems manageable in isolation. Collectively, repetitions can become a major operational expense.

The real cost is not only the time spent performing a task. It is also the context switching, coordination, delay, and uncertainty associated with doing it again.

Repetition can also disguise the original failure.

Once a team becomes accustomed to a workaround, the workaround starts to feel like the process itself.

Staff members may become exceptionally skilled at compensating for a flawed system.

That competence is valuable, but it can conceal the need for redesign.

A practice may celebrate how quickly its billing team clears a backlog without asking why the backlog repeatedly returns.

It may reward employees for resolving denials without examining whether certain categories of denials are preventable.

It may add staff to manage growing administrative volume when some of that volume is generated by avoidable rework.

To be clear, not every denial is preventable. Payer decisions, coverage limitations, medical necessity requirements, and changing policies all contribute to reimbursement complexity.

But the appropriate response is not to treat every denial as an isolated event.

Recurring work should trigger a question about the process that produces it.

If the same issue appears every week, the organization should investigate whether it has a recurring cause.

Otherwise, it risks building an entire operating model around cleaning up after itself.

A Practical Framework: Find the Friction Before Buying the Fix

What can an independent practice do differently?

Start with observation, not procurement.

Before purchasing another platform or launching a broad automation project, identify where work gets delayed, repeated, or returned for correction.

Step 1: Map one process from beginning to end

Choose a process with visible friction, such as insurance verification, authorization, claim submission, denial management, or payment posting.

Document the actual steps—not the steps the procedure manual says should happen.

Ask staff to explain what they really do, including workarounds, spreadsheets, emails, and repeated data entry.

The objective is not to assign blame. It is to understand how work moves through the practice.

Step 2: Identify where information becomes incomplete or unreliable

At each handoff, ask:

• What information is required?
• Where is it first captured?
• Who verifies it?
• Who owns resolving a discrepancy?
• Can incomplete work move forward?
• How does the next person know what has already happened?

Look for points where information is lost, duplicated, delayed, or reinterpreted.

Step 3: Separate preventable errors from unavoidable exceptions

Not every problem has the same cause.

A missing field may be preventable through better capture and validation. A payer policy change may require an updated workflow. An ambiguous clinical situation may require physician judgment.

Classifying problems correctly prevents the practice from expecting automation to solve issues that require human decisions.

Step 4: Measure the baseline

Before implementing a change, record current performance.

Depending on the workflow, useful measures may include:

• First-pass claim acceptance rate.
• Denial rate and denial reasons.
• Time from service to claim submission.
• Days in accounts receivable.
• Percentage of claims requiring manual intervention.
• Staff time spent on rework.
• Number of repeated follow-ups per unresolved issue.
• Time from denial to resolution.
• Payment delays associated with missing information.

Use consistent definitions and compare equivalent periods. Where possible, segment results by payer, service line, or denial category.

Do not measure everything simply because a dashboard allows it.

Measure what helps you understand the problem.

Step 5: Fix the simplest upstream cause first

A process redesign may be more effective than a new platform.

Examples include clarifying ownership, standardizing required fields, removing duplicate entry, improving staff training, or creating a reliable exception-handling process.

If a technology solution is needed, choose one that addresses the identified cause rather than simply adding another place to monitor the consequences.

Step 6: Evaluate the result—not the sales demonstration

After implementation, compare results with the baseline.

Did rework decline? Did the time to resolve exceptions improve? Did staff spend less time on repetitive follow-up? Did the improvement persist after the initial rollout?

Include implementation, maintenance, training, and human-oversight costs.

A tool that saves time in one department but creates more work in another may simply move the burden.

The goal is not to automate a task and declare victory. The goal is to improve the performance of the entire process.

Metrics That Matter—and Metrics That Can Mislead

Metrics are essential, but they can become counterproductive when treated as goals without context.

Claim submission volume: More claims submitted may indicate greater throughput. It does not, by itself, tell you whether claims are accurate, accepted, or paid.

Denial rate: A lower denial rate can be a positive signal, but the reasons for denials matter. Practices should distinguish preventable errors from denials arising from coverage, policy, or medical necessity disputes.

Days in accounts receivable: This helps track collection performance, but it should be interpreted alongside payer mix, claim complexity, aging categories, and the practice’s operating context.

Staff productivity: More tasks completed per employee may suggest greater capacity. But if employees are rushing, overlooking exceptions, or generating downstream corrections, apparent productivity can be misleading.

Automation rate: A high percentage of automated tasks does not guarantee better outcomes. Some tasks require human judgment, and some automated steps may be unnecessary in the first place.

A useful measurement system balances speed, quality, financial performance, and the human effort required to achieve them.

One question should accompany every efficiency metric:

What happened to the work that used to follow this step?

If a process becomes faster but creates more downstream correction, the organization may have improved one number while worsening the system.

Five Healthcare Efficiency Myths Worth Challenging

Myth 1: “If we buy better software, our billing problems will disappear.”

Reality: Software can improve processes, but it cannot automatically correct every upstream data problem, unclear responsibility, or flawed workflow.

Technology works best when its role in the process is well defined.

Myth 2: “Our staff just needs more training.”

Reality: Training matters. But recurring problems can also indicate poor system design, confusing interfaces, inconsistent rules, or unclear ownership.

If competent employees repeatedly encounter the same obstacle, investigate the obstacle before assuming the employees are the problem.

Myth 3: “AI will eliminate medical billing errors.”

Reality: AI can identify patterns, assist with documentation, and automate selected tasks. It can also produce incorrect suggestions or miss important context.

Its output requires appropriate validation, oversight, privacy safeguards, and accountability.

Myth 4: “More automation always means lower costs.”

Reality: Automation can introduce implementation expenses, monitoring requirements, exceptions, integration challenges, and new failure modes.

The meaningful question is whether the total cost of completing the process decreases while quality and compliance are maintained.

Myth 5: “A submitted claim means the job is done.”

Reality: Submission is a milestone, not the final financial outcome.

The claim may still require adjudication, correction, appeal, or follow-up. Practices need visibility into the full lifecycle.

The broader lesson is simple: do not confuse an operational milestone with the outcome it is meant to produce.

The Ethical Question: What Should We Automate?

Not every administrative task should be handed to AI without careful consideration.

Medical billing involves sensitive patient information, financial consequences, payer rules, and decisions that may require professional judgment.

A responsible automation strategy should address several questions.

Privacy: Is patient information being handled through secure, appropriately governed systems?

Accuracy: Can staff verify outputs, identify errors, and correct them before harm occurs?

Transparency: Can the practice understand why a system flagged a claim or recommended an action?

Accountability: Who is responsible when an automated process makes an error or fails to identify one?

Human oversight: Which decisions can be automated safely, and which require review by qualified personnel?

Compliance: Does the workflow comply with applicable privacy, billing, documentation, payer, and contractual requirements?

These are not obstacles to innovation. They are conditions for trustworthy innovation.

The aim should not be to remove humans from every process. It should be to remove unnecessary work while preserving human judgment where it matters.

The smartest automation is not necessarily the automation that does the most.

It is the automation that knows its role—and knows when to stop.

What Wren Roberts’ Story Can Teach Us About Systems

There is an important boundary to this comparison.

Wren’s medical journey is a story about a child with rare and serious congenital conditions, her family’s commitment, and a medical team responding to her individual circumstances. It is not evidence that billing technology can improve rare-disease outcomes.

The connection is a principle, not a clinical equivalence.

In practice operations, we also make assumptions.

We assume the process is working because the dashboard is green.

We assume the claim is progressing because it was submitted.

We assume the problem is resolved because the task was closed.

We assume the new technology is effective because staff have adopted it.

But reality deserves to be checked.

A claim may be submitted and remain unpaid. A task may be closed while a discrepancy persists. A workflow may be automated while creating more exceptions elsewhere.

The right response is not to abandon measurement or planning. It is to treat them as tools for understanding reality, rather than substitutes for it.

A metric is a signal. It is not the whole story.

Wren’s story also reminds us why the human being must remain at the center of the system.

In a medical practice, the patient is more than a diagnosis. The physician is more than a productivity measure. The billing specialist is more than a count of processed claims.

People are the reason the system exists.

When the system begins optimizing itself at their expense, it is time to reconsider what progress means.

The Future of Medical Billing: Less Rework, More Reliability

The future of medical billing should not be defined solely by faster claims processing or more sophisticated AI.

It should be defined by whether practices can make the entire process more reliable.

That means improving the quality of information at the point of capture, making responsibilities visible, identifying exceptions early, and reducing avoidable handoffs.

It also means designing technology around the realities of small and medium-sized physician-owned practices.

These practices do not need complexity for its own sake. They need tools that address real operational constraints without requiring them to build a technology department just to manage the technology.

For AI-powered billing systems, the opportunity is to help practices move beyond reactive correction.

Instead of focusing only on what happens after a claim fails, systems can help identify patterns, surface missing information, support consistent workflows, and give staff better visibility into unresolved issues.

But the outcome must be evaluated honestly.

If a tool saves five minutes on claim submission while creating ten minutes of additional review, it has not delivered the promised efficiency.

If it reduces manual work but increases errors, it has not delivered meaningful improvement.

If it creates a beautiful dashboard while the underlying problem remains unresolved, it has improved the presentation—not necessarily the operation.

The long-term opportunity is to make billing more predictable, reduce avoidable administrative work, and allow people to focus their attention where it produces the greatest value.

That is a more useful definition of innovation than simply adding AI to another step.

Where OnnX Fits Into This Conversation

As a physician-entrepreneur and founder of OnnX, an AI-powered medical billing SaaS designed for small and medium-sized physician-owned clinics, I believe the conversation should start earlier in the process.

We should ask not only how to process a claim more efficiently, but also why the information required to process it was incomplete or inconsistent in the first place.

We should ask whether a task needs to exist, whether a handoff can be simplified, and whether a recurring error can be prevented before it creates another round of administrative work.

That is the thinking behind OnnX’s focus on the relationship between data quality and billing operations.

The objective is not to replace the judgment of physicians or the expertise of billing professionals. It is to reduce unnecessary friction and repetitive work so people can concentrate on the decisions and responsibilities that genuinely require them.

This is a problem to investigate, validate, and solve with practices—not a claim that every billing problem can be eliminated by AI.

The important work begins with understanding what clinic owners and their teams actually experience, where the process breaks down, and which changes would make a measurable difference.

That is how a useful product should develop: from real operational problems, not from the assumption that every problem needs another dashboard.

Three Questions Every Practice Owner Should Ask

If you lead an independent medical practice, consider these questions at your next operations meeting:

1. What work does our team repeat most often—and what causes it to repeat?
2. Which problems do we discover only after they have already created additional work?
3. If we could eliminate one administrative step entirely, rather than automate it, which step would we choose?

The answers may reveal opportunities that a software demonstration would never uncover.

They may also reveal that the solution is simpler than expected: clearer ownership, better information capture, fewer duplicate entries, or a more reliable process for handling exceptions.

Or they may show that technology is exactly what the practice needs.

The point is to let the problem determine the solution—not the other way around.

The Final Thought: Stop Celebrating Work That Should Never Have Existed

Wren Michelle Roberts’ story reminds us that the reality in front of us deserves attention, even when it differs from what we expected.

For healthcare operations, that principle leads to an uncomfortable possibility:

Some of the work we celebrate automating may be work we should have been trying to eliminate.

We can build faster claim processors, smarter denial dashboards, more sophisticated documentation tools, and AI agents capable of completing increasingly complex tasks.

Those advances may be valuable.

But if the same missing information continues to trigger the same corrections, if the same handoffs continue to lose the same details, and if the same employees continue to chase the same unresolved problems, we should ask whether we are making meaningful progress.

Or simply becoming more efficient at compensating for inefficiency.

The future of healthcare innovation should not be measured only by what technology can do. It should also be measured by how much unnecessary work people no longer have to do.

For physicians, that could mean more attention for patients and less time spent untangling administrative problems.

For billing teams, it could mean fewer repetitive corrections and clearer responsibility.

For independent practices, it could mean more predictable operations and less time spent managing avoidable friction.

And for healthcare technology companies, it means accepting a more demanding standard of success: not merely automating activity, but improving outcomes that matter.

Perhaps the most useful question for the next generation of healthcare technology is not:

What can we automate next?

It is:

What should never have required so much work in the first place?

Join the conversation

Physicians and clinic owners: What is the most frustrating recurring administrative task in your practice—one that everyone has learned to tolerate but nobody has truly solved?

Share your experience in the comments. Your answer may reveal a bigger systems problem than another software feature can address.

If this perspective resonates with you, share it with a physician, practice manager, or healthcare operator who is rethinking what efficiency should mean.

For additional practice-improvement resources, see the Featured section of my LinkedIn profile, where you can access a free download with no signup required.

Connect with Dr. Cham on LinkedIn to learn more.

About the Author

Dr. Daniel Cham is a physician, medical consultant, and physician-entrepreneur focused on healthcare technology, practice operations, and medical billing. He is the founder of OnnX, an AI-powered medical billing SaaS focused on reducing unnecessary administrative friction for small and medium-sized physician-owned clinics.

• Website: drdanielcham.com
• Spotify: Listen to the podcast
• YouTube: Watch and subscribe
• X: Follow updates
• Facebook: Join the conversation
• Substack: Read more insights
• TikTok: Explore short-form content
• Instagram: Follow on Instagram

Disclaimer: This article discusses healthcare operations and technology, not individualized medical, legal, billing, or reimbursement advice. Wren Michelle Roberts’ story is based on published reporting; her medical journey should not be interpreted as a guarantee of outcomes for other patients. Any technology used in healthcare should be evaluated for accuracy, privacy, security, compliance, and appropriate human oversight.

References and further reading

1. Vargas, Ramon Antonio. The Guardian. “US infant defies bleak odds from rare congenital trait: ‘Don’t quit before your child quits’”, October 8, 2026.
2. Brott, Thom. BayToday. “The physicians we keep matter as much as the physicians we recruit”, October 7, 2026.
3. American Medical Association. “Physicians’ greatest use for AI? Cutting administrative burdens”, March 20, 2025.
4. Commonwealth Fund. “Administrative Burden in Primary Care: Causes, Potential Solutions”, October 2, 2025.

#HealthcareInnovation #MedicalBilling #PhysicianBurnout #HealthcareAI #RevenueCycleManagement #PracticeManagement #HealthTech #IndependentPractice #HealthcareAdministration #ArtificialIntelligence