Sunday, August 9, 2026

Where Should We Die? The South Korean Story That Exposes What Healthcare Keeps Getting Wrong

We built a healthcare system that can move a patient almost anywhere. The harder question is whether it can carry the patient's story with them.



“How people die remains in the memory of those who live on.” — Dame Cicely Saunders (1918–2005) was a British physician, nurse, social worker, and writer who is widely regarded as the founder of the modern hospice movement.


A patient is lying in a hospital bed in Seoul.

The monitors are working.

The laboratory results are coming in.

The physicians are doing what physicians are trained to do.

The nurses are watching closely.

The family is waiting.

From the outside, it looks like healthcare is working.

But there is another question nobody can answer with a monitor:

Is this where the patient wanted to be?

That question sits underneath recent reporting from South Korea about older adults, long-term-care facilities and hospital deaths.

It is easy to read the story as a debate about hospitals.

It is not.

It is a story about what happens when healthcare becomes better at treating the body than understanding the person.

And that should concern every physician, clinic owner and healthcare entrepreneur.

Because the same thing happens much earlier in the patient's journey.

The patient's story gets fragmented.

The physician sees one piece.

The specialist sees another.

The hospital sees another.

The payer sees another.

The billing team sees another.

And somewhere along the way, everyone has data.

But nobody has the whole story.

That may be one of healthcare's most expensive contradictions.


The Provocative Question

What if healthcare's biggest problem isn't a lack of information?

What if it is that we have too much information in too many places, with too little meaning attached to it?

Think about the modern medical record.

Diagnosis.

Medication.

Lab.

Imaging.

Procedure.

Referral.

Authorization.

Progress note.

Discharge summary.

Claim.

Denial.

Appeal.

Message.

Care plan.

Advance directive.

We have never generated more healthcare data.

Yet patients still say:

"Why do I have to tell my story again?"

Physicians still say:

"Why don't they already know this?"

Nurses still say:

"I can't find what I need."

Billing teams still say:

"The documentation doesn't support the claim."

Patients are confused.

Physicians are exhausted.

Staff are overloaded.

Payers are skeptical.

And healthcare organizations respond in the most predictable way possible:

They buy another piece of software.

That is the part I want to challenge.


The South Korean Story

On August 8, 2026, reporting from South Korea examined why many older adults continue to die in hospitals despite the growth of long-term-care facilities. The story highlights the complicated intersection of aging, hospital transfers, care environments and financial incentives. (chosun.com)

South Korea provides a powerful lens because it is confronting rapid population aging and growing demand for long-term care.

But the story is not really about Korea.

It is about us.

Because eventually every healthcare system faces the same question:

When treatment is no longer the only objective, what does good care actually mean?

Maybe the answer is not another intervention.

Maybe it is:

Familiarity.

Comfort.

Family.

Dignity.

Continuity.

Choice.

And a healthcare system capable of remembering what the patient said before the crisis.


The Patient Has Become the Integration Layer

Here is my contrarian take:

The patient should not be the API.

Yet that is often exactly what happens.

The patient carries the information between systems.

"Here's what my doctor told me."

"Here's my medication list."

"This is what the specialist said."

"My daughter has the paperwork."

"I had that test last month."

"The hospital should have the records."

"The nursing facility sent everything."

Everybody assumes somebody else has the information.

So the patient becomes the human middleware.

That is absurd.

We would never design a technology architecture this way.

Yet healthcare routinely does.


The Most Important Person in the Story Has No Name

The current South Korean reporting does not identify one individual patient whose story we can responsibly reproduce.

And that matters.

Because journalism should not manufacture a character simply because a story would be more emotionally powerful.

Instead, the human story exists in the thousands of patients represented by the data.

The older woman.

The older man.

The daughter sitting beside the bed.

The son trying to make a decision.

The nurse who has known the resident for months.

The physician seeing the patient for the first time.

The family asking:

"What would she have wanted?"

That question is more powerful than a celebrity name.

Because someday, almost everyone will be on one side of it.


Seoul National University Hospital Offers a Different Lens

At Seoul National University Hospital, the Palliative Care and Clinical Ethics Center takes a multidisciplinary approach involving physicians, nurses and social workers.

Its work includes symptom management, advance-care planning, family support and coordination with other care organizations. (snuh.org)

That model challenges one of medicine's oldest assumptions:

If we cannot cure the disease, we have less to do.

Actually, there may be more to do.

Control symptoms.

Clarify goals.

Support families.

Address fear.

Discuss choices.

Coordinate care.

Preserve dignity.

Help people understand what happens next.

In other words:

Medicine does not end when cure ends.

Sometimes the definition of medicine simply changes.


The Question We Avoid

Healthcare loves measurable outcomes.

Mortality.

Readmissions.

Length of stay.

Collections.

Denials.

Utilization.

Patient satisfaction.

Productivity.

But how do we measure something less tangible?

Did the healthcare system remember what mattered to the patient?

That is harder.

There is no simple CPT code for it.

No clean dashboard.

No universal KPI.

Yet it may be one of the most important outcomes.


This Is Where Physicians Should Pay Attention

Because this problem does not begin at the hospital.

It begins in ordinary outpatient care.

A patient comes into your clinic.

You listen.

You examine.

You make a decision.

You document.

You order a test.

You refer.

Then the patient leaves.

Your part of the story is over.

Except it isn't.

The next clinician needs to understand what happened.

The patient needs to remember the plan.

The staff needs to process the referral.

The payer may require authorization.

The billing team needs documentation.

The claim needs to represent the encounter.

And eventually someone needs to know whether the patient actually received the intended care.

One clinical encounter can create a chain of consequences.

The note is not the end of the encounter. It is the beginning of the next one.

That is a profound operational distinction.


The Billing Problem Starts Earlier Than We Think

Here is another contrarian idea:

Revenue-cycle management is not primarily a billing problem.

It is a clinical information problem that eventually becomes a billing problem.

Think about a denied claim.

The easy response is:

"Why did the payer deny it?"

The better question is:

"Where did the information become insufficient?"

Maybe eligibility was wrong.

Maybe authorization was missing.

Maybe the diagnosis was incomplete.

Maybe documentation did not support the service.

Maybe the coding did not accurately reflect the encounter.

Maybe payer policy changed.

Maybe the claim was simply wrong.

But the denial is downstream.

By the time you see it, the original problem may be weeks old.

And now someone has to reconstruct the past.

That is expensive.


We Have Built an Industry Around Fixing Yesterday

This is where I think healthcare needs a rethink.

A large part of revenue-cycle management exists to correct problems after they happen.

Denied claim?

Work it.

Missing documentation?

Find it.

Incorrect code?

Correct it.

Authorization issue?

Appeal it.

Eligibility problem?

Verify it.

Underpayment?

Investigate it.

It is necessary work.

But it is still rework.

And rework is a tax.

Healthcare has become remarkably efficient at creating work to fix work.

That may be one of the least discussed problems in modern medical practice.


The Complexity Addiction

Our instinct is usually to add.

Another dashboard.

Another portal.

Another workflow.

Another vendor.

Another AI assistant.

Another notification.

Another queue.

Another integration.

Another login.

At some point, we have to ask:

What if the solution is subtraction?

What if the best technology is the technology that removes three steps rather than adds another capability?

What if the goal isn't to give staff more information?

What if it is to give them less information, but better organized?

That is a different philosophy.


More Data Is Not the Same as Better Data

This may be the most important distinction in healthcare technology.

More data is not necessarily better data.

A physician does not need 10,000 fields.

The physician needs the right information when making a decision.

A biller does not need the entire patient chart.

The biller needs the information necessary to submit an accurate claim.

A specialist does not need every piece of historical noise.

The specialist needs to understand the clinical question.

A family member does not need a 40-page discharge packet.

They need to know:

What happened?

What do we do now?

When do we call someone?

What should we expect?

Information has value only when it helps someone make a better decision.


Data Exchange Is Not Context Exchange

Healthcare loves the word interoperability.

And interoperability matters.

But here is the problem:

Two systems can exchange data perfectly and still fail to communicate.

A specialist can receive a referral.

But does the specialist understand the clinical question?

A hospital can receive a medication list.

But does the team know which medication the patient stopped taking?

A payer can receive a claim.

But does the claim accurately represent the clinical work?

A nursing facility can receive a discharge summary.

But does the receiving nurse know what the family is most worried about?

Data moved. Meaning did not.

That is not true interoperability.


The Family Becomes the Database

When systems fail to connect, families compensate.

They keep folders.

They save screenshots.

They remember medication names.

They photograph discharge instructions.

They call offices.

They repeat the story.

They become experts in a healthcare system they never asked to study.

This is especially painful when the patient is elderly, cognitively impaired or seriously ill.

The family becomes:

historian + coordinator + advocate + translator + project manager.

That is too much responsibility to place on a person who is already worried about losing someone they love.


The Hidden Lesson for Independent Physicians

Large health systems can build entire departments around coordination.

Independent practices cannot.

That changes the technology equation.

The independent physician does not need another complicated enterprise platform.

They need less friction.

Less duplicate entry.

Less searching.

Less manual correction.

Less chasing.

Less uncertainty.

Less rework.

More visibility.

More predictable workflows.

More time with patients.

That is where healthcare technology should earn its place.


What the Numbers Tell Us

The World Health Organization estimates that approximately 56.8 million people worldwide need palliative care each year, including millions of people in the final year of life. WHO also emphasizes that palliative care can improve quality of life for patients and families and can reduce unnecessary healthcare utilization when appropriately integrated. (who.int)

That is not a small population.

It is a warning.

As populations age, healthcare will have to become better at managing not only disease, but complex human journeys through disease.

That requires continuity.

And continuity requires information that survives transitions.


Three Expert Lessons

Expert Lesson #1: Seoul National University Hospital

The multidisciplinary approach at Seoul National University Hospital demonstrates that serious illness cannot be managed by one discipline alone.

Physicians see the disease.

Nurses see the patient.

Social workers see the family and social environment.

Good care requires all three perspectives.

The operational lesson:

Healthcare information should not be designed around one professional's workflow.

It should support the entire care team.

 

Expert Lesson #2: Hyejin Kim and Colleagues

Research examining surrogate decision-making in Korean long-term-care hospitals highlights the difficult role families play when patients cannot make decisions independently. (pubmed.ncbi.nlm.nih.gov)

The lesson:

Patient preference is not a soft variable.

It can become critical clinical information.

If it is not documented clearly, it becomes memory.

And memory becomes unreliable under stress.

 

Expert Lesson #3: World Health Organization

WHO frames palliative care as an approach designed to improve quality of life and relieve suffering for patients and families. It emphasizes integrating palliative care into broader health systems rather than treating it as an isolated service. (who.int)

The broader lesson:

Healthcare should not wait until the crisis to ask what matters.

The conversation should happen before the ambulance arrives.


The Failure We Don't Talk About

Healthcare organizations often celebrate successful interventions.

But what about successful prevention of administrative work?

Nobody throws a party because a denial never happened.

Nobody celebrates the referral that arrived complete.

Nobody gives an award because the physician did not have to correct a claim.

Nobody creates a press release because a patient did not have to repeat their medication list.

But these are successes.

They are invisible successes.

And invisible successes are exactly what good infrastructure creates.


A Better Definition of Efficiency

We usually define efficiency as:

More patients per hour.

I would add another definition:

Fewer unnecessary steps per patient.

That changes the conversation.

If a physician sees 25 patients but creates 100 downstream administrative tasks, was that efficient?

If a clinic collects $1 million but spends enormous staff time correcting preventable errors, was that efficient?

If a patient completes a referral but has to make five phone calls to accomplish it, was that efficient?

If a hospital discharges a patient quickly but the outpatient physician receives incomplete information, was that efficient?

Efficiency cannot be measured only at one point in the system.

A shortcut for one department can become a burden for another.


The Revenue Cycle Is a Clinical Story in Disguise

Every claim tells a story.

A patient arrived.

A problem was identified.

A physician made a decision.

A service occurred.

Something was documented.

A diagnosis supported the service.

A claim represented the encounter.

A payer evaluated it.

Money moved.

When the claim fails, something in that story becomes questionable.

That does not automatically mean the physician did anything wrong.

It may mean the information was incomplete.

Or the workflow broke.

Or the payer interpreted something differently.

Or the system introduced an error.

That is why blindly attacking the denial is often the wrong first move.

Trace the story backward.


The 20-Claim Test

Here is a simple exercise every independent practice can perform.

Take the last 20 denied or delayed claims.

For each one, ask:

Where did the problem begin?

Not where did the claim fail.

Where did the problem begin?

Was it:

Eligibility?

Authorization?

Documentation?

Clinical information?

Coding?

Payer policy?

Workflow?

Data entry?

Technology?

Human error?

Then calculate the percentage in each category.

You may discover that your biggest billing problem isn't billing.

That is the point.


Five Metrics I Would Add

If I were running an independent practice, I would track:

1. First-pass claim rate

How often does the claim leave the practice correctly the first time?

2. Rework rate

How many times does staff touch the same claim or workflow?

3. Documentation exception rate

How often is downstream work delayed because information is missing?

4. Referral completion rate

How often does the intended next step actually happen?

5. Patient repetition rate

How often does the patient have to repeat information already provided?

That last metric is unconventional.

It should not be.


The AI Trap

Now we reach the fashionable part of healthcare.

AI.

Everyone wants to know:

"Where can AI help?"

I would ask a different question.

"Where is the workflow breaking?"

Then:

"Why?"

Then:

"What information is missing?"

Then:

"Can AI actually solve that problem?"

Sometimes yes.

Sometimes no.

AI can summarize a terrible workflow.

It can automate a terrible workflow.

It can accelerate a terrible workflow.

It can even make a terrible workflow harder to understand.

Automation does not cure bad architecture.

It scales it.

That is why healthcare needs better upstream data before it needs another AI layer.


Where OnnX Comes Into the Conversation

This is the thinking behind my work with OnnX.

The goal is not to make physicians become billing experts.

It is the opposite.

The goal is to reduce the amount of administrative energy physicians and clinic staff must spend translating clinical work into downstream revenue-cycle activity.

I believe the opportunity begins upstream:

Capture.

Structure.

Connect.

Validate.

Learn.

The closer we can connect clinical reality with operational and billing workflows, the less downstream reconstruction should be required.

That is the thesis.

Not:

"Let's build another billing tool."

But:

"Let's make the clinical information entering the revenue cycle more usable in the first place."


Why This Matters More for Small Practices

Large organizations can absorb complexity.

Independent practices cannot.

A health system might be able to assign ten people to a workflow.

An independent clinic may have one person doing three jobs.

That means every unnecessary task has an outsized impact.

One missing authorization can consume an hour.

One documentation correction can delay a claim.

One referral problem can generate multiple calls.

One payer rule change can create a new workflow.

Small practices don't need more complexity.

They need leverage.


The Practical Playbook

Step 1: Find your worst bottleneck

Do not start with your favorite technology.

Start with your biggest recurring headache.

 

Step 2: Follow the information

Where is the information created?

Who receives it?

Who changes it?

Who re-enters it?

Where does it disappear?

 

Step 3: Find the first point of failure

The first failure matters more than the final symptom.

A denial may be the symptom.

The real problem may have occurred during scheduling, registration, authorization, documentation or coding.

 

Step 4: Remove one handoff

Every handoff is a potential failure point.

Ask whether the handoff is necessary.

 

Step 5: Standardize the predictable

If the same problem occurs repeatedly, stop treating it as a surprise.

Create a standard workflow.

 

Step 6: Automate carefully

Automate repetitive work.

Do not automate clinical judgment simply because you can.

 

Step 7: Measure the result

Did staff time decrease?

Did rework decrease?

Did first-pass claims increase?

Did referral completion improve?

Did patients repeat themselves less?

If not, the technology did not solve the problem.


Pitfalls to Avoid

Pitfall 1: Buying software before mapping the workflow

Technology can hide a process problem.

Pitfall 2: Measuring revenue without measuring rework

Collections tell you what happened financially.

They do not always tell you why.

Pitfall 3: Assuming documentation means longer notes

It does not.

Clarity beats volume.

Pitfall 4: Treating patients as data sources

Patients are people.

Their preferences are not merely fields.

Pitfall 5: Treating palliative care as failure

Comfort, dignity and symptom management are legitimate clinical goals.

Pitfall 6: Assuming AI will fix everything

AI is a tool.

It is not an operating model.


Ethical Considerations

There is a deeper ethical issue here.

When we turn a patient into data, we risk forgetting that the data represents a human life.

A diagnosis is not merely a code.

A care preference is not merely a checkbox.

A discharge summary is not merely a document.

A claim is not merely a transaction.

Each represents something that happened to a person.

The technology should therefore protect:

Autonomy.

Privacy.

Accuracy.

Continuity.

Human judgment.

And especially when a patient cannot speak for themselves, the system should make it easier—not harder—to understand what they previously expressed.


Legal and Compliance Considerations

Better information continuity can also reduce operational and compliance risk, but documentation is not a substitute for professional judgment or legal advice.

Physicians and practices should document relevant clinical reasoning, patient preferences, consent, treatment decisions and care plans appropriately.

They should also understand applicable requirements involving:

HIPAA and privacy.

Advance directives.

Informed consent.

Payer documentation requirements.

Coding and billing rules.

Medical necessity.

State-specific requirements.

The critical principle is:

Document the clinical reality accurately.

Never create documentation simply because a payer or billing workflow appears to demand a particular clinical conclusion.

The record should reflect what actually happened.


What I Would Change About Healthcare's "Best Practices"

Here are several conventional assumptions I would challenge.

Best practice: Add more technology.

My question: Can we remove a step instead?

 

Best practice: Capture everything.

My question: Can we capture what matters better?

 

Best practice: Build another dashboard.

My question: Who is making the decision, and what do they actually need to see?

 

Best practice: Optimize each department.

My question: What happens to the next department?

 

Best practice: Fix denials faster.

My question: Why are the same denials happening repeatedly?

 

Best practice: Automate documentation.

My question: Are we improving the clinical story or simply generating more text?


The Bigger Contrarian Idea

Maybe healthcare does not have a technology problem.

Maybe it has a translation problem.

We constantly translate:

Patient → clinician.

Clinician → documentation.

Documentation → code.

Code → claim.

Claim → payer.

Payer → payment.

Payment → financial report.

At every translation, meaning can be lost.

The future belongs to systems that preserve meaning across those translations.

That is much harder than simply moving data.

But it is where the value is.


The Future of Healthcare Billing May Begin at the Bedside

This is why I believe the next generation of revenue-cycle innovation will move upstream.

Not because billing departments are unimportant.

Because downstream teams can only work with the information they receive.

The more accurately the clinical story is structured at the beginning, the more predictable the downstream process becomes.

That is the thesis behind OnnX.

Better upstream information.

Less downstream friction.

Fewer manual corrections.

More predictable revenue.

And, ideally:

More physician time spent practicing medicine.


The Future Outlook

Healthcare is becoming increasingly distributed.

Care is moving into:

Homes.

Clinics.

Ambulatory centers.

Long-term-care facilities.

Specialty practices.

Virtual environments.

Community settings.

As the physical healthcare system becomes more distributed, information continuity becomes more important, not less.

The winning healthcare organizations will not necessarily be the ones with the most technology.

They will be the ones that make transitions feel invisible.

The patient should move.

The information should move.

The clinical intent should move.

The financial representation should move.

The human context should move.

Without forcing the patient to carry everything themselves.


The Question I Want Healthcare Leaders to Answer

Imagine your next patient moving through your organization.

At every transition, ask:

Does the next person know what the last person knew?

If the answer is no, you have a workflow problem.

If the answer is sometimes, you have a reliability problem.

If the answer is yes only because an employee remembers, you have a scalability problem.

And if the answer is yes because the system reliably carries the information forward?

You may finally have something resembling true interoperability.


Final Thoughts: Don't Lose the Person Inside the Data

The South Korean story begins with an older person's final journey.

But it ends somewhere much closer to home.

In your clinic.

In your EHR.

In your referral queue.

In your documentation.

In your claims.

In your inbox.

In the hours your staff spend fixing problems that should never have existed.

The lesson is not that hospitals are bad.

It is not that long-term-care facilities are better.

It is not even that technology is failing.

The lesson is simpler.

Healthcare can only care for the person it can understand.

And understanding requires more than data.

It requires context.

It requires continuity.

It requires listening.

It requires remembering.

And sometimes, it requires knowing when the most important thing to preserve is not another laboratory result.

It is the patient's voice.


Three Questions for Physicians and Clinic Owners

Where does your patient's story disappear in your workflow?

What recurring administrative problem are you fixing today that should have been prevented yesterday?

How much of your staff's time is spent translating information instead of caring for patients?

I would genuinely like to hear your answer.

Leave a comment.

Tell me where the friction begins in your practice.

And if this perspective resonates with another physician or clinic owner, repost it.

The conversation should not be about whether healthcare needs more technology.

It should be about whether the technology we already have is helping us preserve what matters.


Get Involved

Healthcare will not become less complicated simply because we call it "digital."

It becomes better when we deliberately remove unnecessary friction.

Raise your hand.

Share what is working—and what is not.

Join the conversation about building healthcare around the patient rather than around the workflow.

If you are a physician, clinic owner, healthcare operator or founder working on these problems, step into the conversation.

Ask better questions.

Challenge comfortable assumptions.

Share your experience.

Help shape what comes next.

Because the next healthcare breakthrough may not be another device, drug or algorithm.

It may be a simpler way for the system to remember what the patient already told us.


About the Author

Dr. Daniel Cham is a physician, medical consultant and healthcare technology entrepreneur focused on the intersection of clinical practice, healthcare operations, medical billing and technology innovation.

As founder of OnnX, he explores how better upstream clinical and operational data can reduce administrative friction, improve revenue-cycle performance and help independent medical practices operate with greater simplicity and predictability.

His perspective comes from looking at healthcare from both sides of the equation: the clinical reality of patient care and the operational reality of running a medical practice.

Connect with Dr. Cham on LinkedIn to learn more.


Disclaimer

This article is provided for general educational purposes and is not intended to constitute medical, legal, coding, compliance or reimbursement advice.

Healthcare requirements vary according to individual circumstances, jurisdiction, payer, clinical setting and applicable law. Physicians, healthcare organizations and other professionals should obtain appropriate professional guidance for specific clinical, legal, regulatory or reimbursement decisions.


Continue the Conversation

Healthcare innovation is not only about what is new.

It is about what becomes easier, clearer and more human because we built it better.

Explore practical perspectives on healthcare operations, medical technology, innovation and the future of physician-owned practices.


Knowledge Drives Progress

The best healthcare ideas do not begin with a product.

They begin with a question.

Why is this harder than it needs to be?

Then another:

Where did the friction begin?

And finally:

What can we change?

Start there.


Free Resource for Physicians and Clinic Owners

If you want practical ideas for improving the revenue cycle in an independent medical practice, visit the Featured section of my LinkedIn profile for a free resource.

No complicated sign-up process.

Just practical information you can explore and apply.

Learn something useful. Question an assumption. Improve one workflow.


Help Move the Conversation Forward

If this article made you think differently about medical billing, clinical documentation, care coordination or healthcare fragmentation, don't keep the conversation to yourself.

Comment with your experience.

Share this article with a physician or clinic owner who deals with these problems every week.

Repost it if you believe healthcare can become simpler without becoming less human.

The patient should never have to carry the burden of connecting a fragmented healthcare system.

The patient should be the person we are connecting it for.


#Healthcare #HealthcareLeadership #PhysicianLeadership #MedicalPractice #IndependentPractice #MedicalBilling #RevenueCycleManagement #HealthcareInnovation #HealthTech #ClinicalDocumentation #PatientCenteredCare #CareCoordination #PalliativeCare #HealthcareOperations #PhysicianEntrepreneur #MedicalTechnology #HealthcareAI #DigitalHealth #IndependentPhysicians #OnnX

 

No comments:

Post a Comment

Where Should We Die? The South Korean Story That Exposes What Healthcare Keeps Getting Wrong

We built a healthcare system that can move a patient almost anywhere. The harder question is whether it can carry the patient's story wi...