One-Stop vs Fragmented: Why Taiwan's 4-Hour Workup Beats America's 12-Week Coordination

The single most under-appreciated structural advantage Taiwan's preventive screening hospitals hold over U.S., Japanese, or European equivalents isn't price, equipment quality, or even physician training. It's workflow density — the deliberate engineering of an entire health workup as a continuous 4-hour experience inside one campus, rather than a 12-week sequence of separate appointments across multiple clinics, each owned by a different specialty, billed under a different code, and stored in a different EHR. This is the "one-stop medical center" model, and once you've experienced it, the U.S. fragmented model feels almost willfully inefficient.

This piece is the operational counterpart to our analysis of full-body MRI access and our overview of preventive gaps in the U.S. — instead of asking what Taiwan does differently, it asks how the workflow itself is engineered, why the U.S. cannot easily replicate it, and what that means for follow-up care over the years that come after the first morning.

How a fragmented system spreads a workup across weeks

Imagine you're a 50-year-old American who wants a comprehensive baseline. Your primary care physician is supportive but limited by what insurance covers. You self-pay for additional services. The realistic timeline:

1. Week 1: PCP appointment to discuss interest in advanced screening. Physical exam, basic labs ordered (CPT 80050 general health panel). Wait for lab results.
2. Week 2: Lab follow-up call. PCP cannot order full-body MRI for asymptomatic screening (no insurance authorization for non-symptomatic imaging); suggests Prenuvo or hospital cash-pay.
3. Week 3-4: Schedule cash-pay full-body MRI at Prenuvo. Wait for the appointment slot. The MRI itself is split across CPT codes 70551 (brain MRI without contrast), 71550 (chest), 74181 (abdomen), and 72148 (lumbar spine) — each a separately billable line item, each individually credentialed.
4. Week 4-5: MRI scan day. Receive AI-summarized report through Prenuvo app several days later. The reading radiologist is a remote teleradiology contractor who never speaks with you and is not responsible for downstream resolution of any incidental finding.
5. Week 5-6: Schedule low-dose lung CT (CPT 71250) at hospital radiology — different facility, different EHR (often Epic on the hospital side, Allscripts at the imaging vendor), separate appointment, separate intake forms, separate radiology report sitting in a separate portal.
6. Week 7: Schedule coronary calcium score at cardiology imaging. Yet another appointment, yet another intake, yet another portal login.
7. Week 8: Schedule DEXA bone density (CPT 77080) at sports medicine clinic or endocrinology imaging. Fourth facility.
8. Week 9: Schedule abdominal ultrasound (CPT 76700) at yet another radiology suite — often this is a 3-week wait for the next available cash-pay slot.
9. Week 9-10: Order specialty biomarker panel through Function Health or independent lab. Blood draw at Quest. Wait for results to land in another portal.
10. Week 11-12: Aggregate everything for review with PCP. PCP allocates 15-20 minutes (insurance-defined) to discuss. The PCP is reading PDFs uploaded by the patient because the imaging facility's PACS does not exchange with the primary care EHR. Integration of findings is, in practice, the patient's job.

Total elapsed: 10-12 weeks. Total facilities visited: 4-5. Total physician time integrating findings: 15-30 minutes. Total cash-pay cost: $4,500-$8,000. Total time off work for the patient: 4-5 days, spread non-contiguously across three months. Total separate billing entities sending statements: typically 6-8.

A representative U.S. fragmentation map

The structural pattern: four to seven separate EHRs, six to eight separate billing entities, no single physician owning the integrated picture. The patient is the integrator by default.

How a one-stop campus does the same in one morning

At our partner Beitou Health Management Hospital — a representative example of Taiwan's one-stop model and a JCI-accredited facility — the same workup unfolds:

1. 7:50 AM: arrive, intake at concierge desk. ID check, fasting blood draw begins immediately.
2. 8:15 AM: full-body 3T MRI begins (60 min); blood vials going to in-house lab in parallel.
3. 9:30 AM: low-dose lung CT (10 min) on the CT suite directly adjacent to MRI — no transit, no second intake.
4. 9:45 AM: coronary calcium score CT (10 min) on the same scanner — protocol pre-loaded by the technologist who already has the patient's chart.
5. 10:00 AM: DEXA in adjacent suite (10 min).
6. 10:30 AM: ultrasound (carotid, thyroid, abdominal) — 25 min, performed by a sonographer who can already see the morning's MRI thumbnails on a shared workstation if needed.
7. 11:00 AM: ECG, echocardiogram if applicable.
8. 11:30 AM: brief intake-physician debrief; lunch on the hospital.
9. Following morning 9:00 AM: 30-minute physician debrief reviewing complete report; recommendations for any follow-up.
10. 7 days later: full digital report delivered (PDF + DICOM compatible with U.S./EU EHRs).

The scheduling logic is the quiet genius of the system. Imaging suites are placed adjacent to each other in a deliberate sequence: blood draw room → MRI → CT → DEXA → ultrasound → cardiac. Patients never travel more than 30 meters between modalities. Each technologist is briefed at 7:30 AM with the day's manifest. The radiology workstation pre-loads the next exam during the prior exam, which collapses the dead time between scans to 2-3 minutes instead of the 30-90 minutes you experience when you arrive at a new facility cold.

Total elapsed: ~24 hours from intake to debrief, 7 days to full report. One facility. One coordinated team. One PACS holding every image. One physician reviewer integrating the findings before the patient walks back in for the debrief. Physician-led integration baked into the schedule rather than offloaded to the patient.

Why fragmentation is structural — five reasons

The U.S. system did not fragment care by accident. It fragmented for five specific structural reasons, each of which would need to be unwound for a true one-morning workup to exist domestically.

1. Billing codes are modality-specific, not workflow-specific. CPT (Current Procedural Terminology) is the language insurance speaks. Each CPT code maps to one procedure on one body region — there is no code for "comprehensive screening morning." Hospitals build operations around what they can bill, and they cannot bill an aggregate experience. So they don't operate one. The closest substitute, "executive physical," is a bundled cash-pay product that exists at Mayo, Cleveland Clinic, and a handful of others — but it is treated as a niche concierge offering, not a core hospital workflow, and it still does not cover full-body MRI.

2. Anti-kickback and Stark Law constraints discourage cross-specialty billing. The Stark Law restricts physician referrals across financially-related entities, and the federal Anti-Kickback Statute restricts revenue-sharing arrangements that look like incentives to refer. The legal effect is to push specialties to operate as separate billing entities even within the same physical hospital. The cardiologist owns the calcium score, the radiologist owns the MRI, the orthopedic group owns the DEXA — and integrating those into a single bundled product creates legal exposure that few institutions choose to take on without a dedicated compliance structure.

3. EHR interoperability is genuinely poor, even within the same metro. Two hospitals down the street from each other, one running Epic and one running Cerner, often cannot reliably exchange imaging or notes. HL7 FHIR has improved this in theory but the practical reality is that imaging in particular still requires CD/DVD transfer, fax, or patient portal upload between institutions. A workflow that depends on cross-institution data flow is a workflow that breaks regularly.

4. Specialty ownership of body regions is deeply institutionalized. The brain MRI is read by a neuroradiologist. The cardiac CT by a cardiac radiologist. Abdominal imaging by abdominal/GU radiologists. DEXA reports often by an endocrinologist. Each subspecialty has its own credentialing, billing, and reporting culture. Building a workflow that runs all of these on one patient on one morning requires those subspecialties to share a queue and a reporting template — which they generally do not.

5. Insurance payment categories are pre-authorization gated. Even cash-pay patients run into the structural inheritance: the imaging center's protocols, scheduling templates, and intake forms were all built for insurance-authorized workflows where each scan is independently approved. Re-engineering them for a bundled cash-pay morning is technically possible but commercially small relative to the insurance volume, so it rarely happens.

Taiwan's system avoided all five of these. The NHI structure pre-bundled outpatient pricing in a way that made adjacency natural. The cash-pay tier could engineer experience without billing-code constraints. EHRs are mostly hospital-built and integrated by default. Specialty silos exist but are softer because medical training in Taiwan emphasizes generalist screening rotation. And there is no insurance pre-authorization layer for the cash-pay product. The result is a workflow that the U.S. cannot easily import — not because U.S. hospitals are less capable, but because the structural layers of the U.S. system point in the opposite direction.

"I wasn't sure I needed a Taiwan trip for screening. Then I added up the appointments I'd need to coordinate at home — MRI here, CT there, blood draw somewhere else, results stitched together by me — and realized I was paying for fragmentation. The Taipei morning was honestly the most efficient health appointment I've ever had." — Carlos M., 47, software architect, Austin, Texas

Forward, One Medical Plus, and the U.S. concierge medicine response

The U.S. has not been blind to fragmentation. A generation of concierge and membership-medicine startups has tried to integrate primary care into a smoother experience — Forward, One Medical (now part of Amazon), Crossover Health, Parsley Health, and Function Health each address pieces of the problem. Their integration scope is worth understanding precisely.

Forward built proprietary clinics with a body composition scanner, EKG, blood draw, and an in-room display that walked patients through their own data. They closed their consumer offering in late 2024 after struggling to scale the unit economics. What Forward integrated well: primary care visits, basic labs, body composition. What they never integrated: full-body MRI, low-dose lung CT, coronary calcium score, DEXA. The capital intensity of imaging was outside their model.

One Medical Plus (since the Amazon acquisition) offers same-day primary care and coordinated specialty referrals. Imaging is still external — they refer to partner radiology networks, which means the patient still has to schedule, travel to, and integrate findings from a separate facility. The integration improves the primary-care experience without changing the imaging fragmentation.

Crossover Health serves employer populations with on-site or near-site clinics that bundle primary care, behavioral health, physical therapy, and basic labs. Imaging beyond ultrasound is referred out. The model improves the front door of the system but leaves the back end — the imaging stack that costs $1M+ per modality to capitalize — untouched.

Function Health is the closest U.S. analog for the labs side: 100+ biomarker subscription with a clinician review layer. It does not include imaging at all. It is, in a real sense, the labs sub-component of a Taiwan workup, sold as a standalone product because the imaging side is structurally too expensive to bundle in.

The pattern is consistent. U.S. integration startups own primary care and labs because those are operationally light. They do not own the imaging stack because each modality requires a $1-3M capital investment and a credentialed reading workforce — which is exactly the bundle Taiwan's hospitals already operate at scale because the public-system installed base lowered the marginal cost. Until a U.S. integrator capitalizes its own imaging suite, the one-morning workup will not exist domestically.

JCI accreditation and what "one-stop" actually requires

Joint Commission International (JCI) is the global standard for hospital accreditation, and several Taiwan health management hospitals — including Beitou and a number of our partners — carry it. JCI is not a marketing badge. It is an audit framework with roughly 1,200 measurable elements across 16 chapters, including:

• International Patient Safety Goals — patient identification, medication safety, surgical safety, infection prevention.
• Access to Care and Continuity of Care — explicitly evaluates whether the patient experiences a coordinated path through the institution. A fragmented intake-to-debrief process fails this audit.
• Assessment of Patients — requires that each modality's findings are assembled into a unified record before the integrating physician's review.
• Care of Patients — requires documented protocols for how findings flow between specialties.
• Management of Information — requires a single integrated medical record. A facility running three different EHRs internally would struggle to pass.
• Facility Management and Safety — covers the physical adjacency of imaging suites, emergency response, equipment maintenance.

The "Access to Care" and "Management of Information" chapters in particular align almost perfectly with what makes a one-stop workup possible. JCI didn't invent the workflow, but the accreditation process actively rewards integration and penalizes fragmentation. A Taiwan health-management hospital pursuing JCI is being audited, every three years, on whether the one-morning model actually works as designed. This is part of why patients can rely on the consistency of the experience between visits.

The follow-up implication — single record vs scattered baseline

One-stop also means one written record, and the implications compound over years. When a finding requires follow-up — say a hepatic lesion that warrants 6-month MRI re-imaging, or a 4mm pulmonary nodule that Fleischner guidelines flag for repeat at 12 months — the entire baseline data is in one institution and one PACS. Comparison imaging the next year is a matter of pulling up the prior study side-by-side on the same workstation. The same radiologist (or one of a small reading team) can review both. Subtle interval change is visible. Confidence in the read is high.

Compare this to the U.S. follow-up scenario. Last year's MRI is at Prenuvo. This year's is being scheduled at a hospital. The hospital radiologist asks the patient for the prior study. The patient logs into the Prenuvo portal, downloads a DICOM ZIP, uploads it to the hospital portal — which often rejects the format or the file size. The patient ends up burning a CD at a kiosk. The CD arrives the day of the new scan, gets imported into the hospital PACS in a low-resolution wrapper that strips metadata. The radiologist now has two studies that are technically comparable but practically uncomfortable to compare. The interval-change confidence drops. Often a third "definitive" scan gets ordered to settle the question.

This is the structural cost of fragmentation that almost nobody prices into the original decision. The first workup is the easy one. The fifth workup, on a returning 55-year-old who is now five years into surveillance for an incidentaloma found at age 50, is where one PACS versus five PACS becomes materially different medicine.

The "incidentaloma coordination problem" — what to do about an ambiguous finding — is also structurally different in the two systems. In the U.S., the reading radiologist is a transactional contractor with no responsibility for downstream resolution. They flag the finding, recommend "clinical correlation," and the buck moves to a primary care physician who is not equipped to clinically correlate a 7mm liver lesion. In Taiwan's one-stop model, the integrating physician who reviews the report at the morning-after debrief is the person who explicitly owns the follow-up decision and books it before the patient leaves. This is not a small difference.

Patient persona narratives — three cases

Case 1: The Texas software architect comparing the two paths. Carlos, 47, lives in Austin and earns enough to cash-pay either route. He runs the math: at home he is looking at 12 weeks, four facilities, eight portals, 4-5 days off work spread across the quarter, and an estimated $6,800 in cash-pay charges. The Taiwan route is one morning of clinical time, one debrief the next morning, three days of total trip time including flights, $3,200 for the screening package, and roughly $1,800 in travel. He picks Taiwan, and the deciding factor is not price — it is the prospect of not having to be his own care coordinator for three months.

Case 2: The post-cancer patient on surveillance. Maria, 58, finished treatment for stage I breast cancer four years ago and is on annual full-body imaging surveillance. Her first three years of imaging were done at three different U.S. facilities because her insurance steered each year toward whichever in-network center had availability. Her oncologist spent the better part of an hour each year hunting down prior images. She switched to a Taiwan annual workup in year four, and now every annual scan since then sits in the same PACS. The interval comparison her oncologist now does in five minutes used to take a week of administrative chasing. The continuity of record is the entire point for her.

Case 3: The couple doing a first joint baseline. James (51) and Priya (49), both executives, decided to do their first comprehensive baseline together. They considered Mayo's Executive Health Program (3 days, $9,000 each, no full-body MRI included) versus a Taiwan one-stop package (1 morning each, $3,400 each, full-body MRI included, plus a 4-day combined trip). They picked Taiwan. What surprised them was not the price difference but the fact that they walked out at noon, had lunch together, and spent the afternoon at a museum. The compression of the workup into a single morning let them treat the trip as a couple's experience rather than a medical chore.

What this means for the future of preventive medicine

Workflow density is becoming a differentiated product in its own right, separate from any individual modality. Function Health, Prenuvo, Ezra, Fountain Life, and the executive physical programs at Mayo and Cleveland Clinic are each excellent at one slice of the workup, but none of them are integrators. They are vertical specialists in a market that fragmented along vertical lines, and being a vertical specialist is a structurally easier business than being an integrator. Integration requires owning the imaging capital, the lab capital, the clinician workforce, and the IT layer simultaneously — which is a hospital-scale undertaking, not a startup-scale one.

Taiwan's advantage is that this hospital-scale undertaking is already paid for. The MRI suites, CT suites, DEXA, ultrasound, and lab were capitalized through the NHI installed base over the past two decades. The cash-pay tier inherits that capital base and operates the integrated workflow on top of it. Reproducing this from scratch in the U.S. would require billions of dollars of imaging capital plus a regulatory environment that does not exist. This is not a gap that closes in 5-10 years.

What the U.S. can do is close pieces of it. Function Health is closing the labs piece. Prenuvo is closing the imaging-as-a-product piece. One Medical is closing the primary-care piece. None of them have closed the integration piece, and the integration piece is exactly what Taiwan sells. For now, and for the foreseeable future, an American who wants the integrated experience will keep finding their flight to Taipei pencils out — not because Taiwan is cheaper or technologically ahead, but because the workflow itself is the product.

For more on the technology layer that sits underneath this workflow, see our companion pieces on Taiwan's 3T MRI installed base and how AI augments the reading layer. To explore concrete packages built around this workflow, see our screening services and partner hospitals.

Sources & Further Reading

• USPSTF — U.S. Preventive Services Task Force recommendations
• American Cancer Society — Cancer Facts & Figures
• NCI SEER — Surveillance, Epidemiology, and End Results Program
• CDC — Cancer prevention & screening
• American Heart Association — heart attack risk
• American College of Cardiology — clinical guidelines

Frequently asked questions