Hypersphere Health, Inc.

The architecture was built
around the billing code.
Not the person.

Healthcare's most persistent failures have been documented for thirty years. Preventable admissions. Avoidable readmissions. Undertreated complexity. The clinical knowledge to close these gaps has always existed. The data architecture has not.

Patent-Pending HPDM™ Architecture  ·  Provisional Filed May 5, 2026  ·  Hypersphere Health, Inc.

Thirty years of data.
The same failures.

These are not projections or estimates. They are published findings from federal agencies and peer-reviewed research — the documented cost of an architectural problem that incremental improvements have not solved.

50%
of total US healthcare expenditure is generated by the top 5% of patients — the high-complexity population that additive risk models consistently underscore.
AHRQ Medical Expenditure Panel Survey (MEPS) — consistent finding across multiple survey years
30–55%
of health outcomes are attributable to social determinants — housing, food security, community environment — that are invisible to claims-based data systems.
WHO Commission on Social Determinants of Health (2008) · Affirmed by CDC, National Academies (NAM)
15–20%
30-day readmission rate for high-complexity patients — persistent despite a decade of CMS financial penalties and focused clinical attention.
CMS Hospital Readmissions Reduction Program (HRRP) — published annual data, multiple years

In 2001, the Institute of Medicine published Crossing the Quality Chasm — the most cited document in modern American healthcare policy. Its finding was not a lack of clinical knowledge or capable technology.

Its finding was structural: the delivery system was not designed to deliver the care that clinical science had already made possible.

"The American health care delivery system is in need of fundamental change... The current system cannot do the job... trying harder will not work. Changing systems of care will." Institute of Medicine · Crossing the Quality Chasm · 2001

A decade later, the IOM's Best Care at Lower Cost (2012) documented that the conditions from 2001 had not fundamentally changed — estimating $750 billion in annual healthcare waste attributable to fragmentation, inefficiency, and missed prevention.

The Medicare Payment Advisory Commission has documented in multiple annual March Reports that Medicare Advantage plans systematically underpay for their highest-complexity members — the precise population where additive risk scoring fails most severely.

The HCC v28 model assigns risk scores additively — each condition weighted independently, as if the others do not exist. This methodology is appropriate for its designed purpose: actuarial population-level payment adjustment.

It was never designed as a clinical intelligence tool. It was never intended to capture how conditions interact. And it cannot be patched to do so — the limitation is architectural, not algorithmic.

MedPAC has consistently documented that the HCC model underpredicts costs for the sickest beneficiaries — particularly those with complex comorbidity combinations the additive architecture cannot resolve. Medicare Payment Advisory Commission · March Reports · Multiple years

Why incremental improvement
cannot close the gap.

A better algorithm running on the wrong architecture produces better predictions of the wrong thing. The problem is not the algorithm. The problem is the organizing unit — the billing code instead of the person.

Additive Architecture — Current Standard

HCC v28 and Every System Built on It

Conditions are scored independently. Each diagnosis receives a risk weight. The patient's total score is the sum of those weights.

Social determinants are not captured. Behavioral factors are not in the model. Functional status is not a variable. Comorbidity interactions are not computed.

The patient is not the organizing unit. The billing code is.

Patient score = Condition A weight
+ Condition B weight
+ Condition C weight
= additive total

Interaction effects: NOT MODELED
HPDM™ Architecture — Patent-Pending

The Person as the Organizing Unit

The patient is placed at the center. All data — clinical, social, behavioral, biological — is organized as an attribute of that whole person, classified by mutability.

Condition interactions are computed multiplicatively across 65 validated comorbidity pairs and 29,403 Z-axis interaction paths.

The score reflects the patient's actual clinical reality — not an administrative approximation of it.

Patient score = f(Shell 0, Shell 1, Shell 2, Shell 3)
× Z-axis comorbidity interaction multipliers
across all active condition combinations

Interaction effects: COMPUTED
"Trying harder with the existing architecture will not produce the outcomes the evidence shows are achievable. Only a different architecture — one that starts with the person as the organizing unit — can close the gap the literature has documented for thirty years." The Architecture Problem — Hypersphere Health, Inc. Working Paper, May 2026

Patent-Pending · Filed May 5, 2026

The Hyper-Spherical Patient
Data Model.

HPDM organizes every data element about a patient as an attribute of that patient — classified by how mutable it is and what domain of their life it reflects. Four shells. Three scoring axes. One whole-person clinical profile.

Shell 0
Immutable Biological
Permanent Baseline
Genetic factors, birth history, sex, race/ethnicity, congenital conditions. These do not change. They form the permanent risk floor all other shells modify.
Shell 1
Near-Immutable Environmental
Social Determinants
Geography, housing stability, community environment, SDOH. Change slowly and constrain what clinical interventions in Shells 2 and 3 can achieve.
Shell 2
Semi-Controllable Personal
Behavioral & Cultural
Health literacy, medication adherence, dietary behavior, lifestyle. Can be influenced by care interventions but not controlled.
Shell 3
Interactable Clinical
Active Clinical Layer
Active diagnoses, medications, procedures, utilization, care gaps, lab results, vitals, RAF-relevant coding. The only layer traditional risk models see.
The Z-Axis — The Differentiator

Multiplicative Comorbidity Interaction Engine

Where HCC v28 adds condition weights independently, HPDM's Z-axis computes interaction effects between conditions — the same multiplicative compounding that clinical literature has documented for decades and that additive models structurally cannot capture.

A patient with diabetes, chronic kidney disease, and heart failure does not carry three independent risks. Each condition accelerates and modifies the progression of the others. A patient with 12 active conditions generates 66 pairwise interaction evaluations under HPDM's Z-axis — against 12 independent assessments in an additive model.

221
Condition library
59
Validated clinical instruments
65
Validated comorbidity pairs
29,403
Z-axis interaction paths
391,702,652
Possible assessment panels

The outcomes your organization
cannot currently achieve.

These are not projections. They are outcomes documented in peer-reviewed literature as achievable — and documented as persistently unachieved with current data architectures. HPDM is the architectural change that makes the difference.

RAF Accuracy
HPDM identifies conditions that are clinically present and documentable but absent from current HCC coding — through complete, accurate documentation of what the clinical record supports.
RAF delta validated at +0.225 to +0.238 across synthetic cohort analyses · Connecticut CT RHT cohort: +0.238 locked figure
Preventable Admissions
HPDM surfaces the patients whose clinical trajectories are deteriorating before a crisis occurs — the diabetic patient with accelerating CKD progression, the heart failure patient whose adherence is compromised by social circumstances.
AHRQ PQI: substantial proportion of admissions potentially preventable with appropriate outpatient management
Avoidable Readmissions
HPDM's whole-person profile identifies the high-complexity patients whose discharge plans are likely to fail — before they fail. Social determinants, behavioral factors, and comorbidity interactions are visible together.
CMS HRRP: 15–20% 30-day readmission rate for high-complexity patients — persistent despite decade of focused intervention
Care Gap Closure
HPDM generates individualized clinical action protocols from the whole-person profile — not population-average care pathways. Every gap identified has an architectural basis, not a statistical one.
510 KPIs tracked across all six activation layers · Protocol generation from Shell 0–3 integrated profile
SDOH Integration
Social determinants are not a separate module added to HPDM — they are Shell 1 of the architecture. They modify the clinical scoring of every other shell. A care plan that ignores Shell 1 will consistently underperform.
WHO: 30–55% of health outcomes attributable to SDOH · Shell 1 captures what no claims-based model can see
VBC Contract Performance
Value-based contracts require the ability to demonstrate measurable outcomes improvement. HPDM's Delta Analytics Engine tracks prediction vs. realized outcomes — producing the evidence base that VBC contracts require.
Six-layer activation model: Layer 1 through VBC All-Payer Shared Savings · $36–72M/yr at coalition scale

Every engagement begins
at the lowest meaningful entry point.

We do not ask for trust before we earn it. Tier Zero is a fixed-fee assessment against a synthetic patient cohort — no integration required, no PHI, no long-term commitment. You see what HPDM finds. You decide what to do next.

Tier Zero — Start Here
Synthetic Cohort Assessment
$15,000 – $25,000 · Fixed Fee · 60–90 Days
Named market analysis. Synthetic cohort build from public Tier A data sources. Cohort Intelligence Brief — Phase Zero analytical output. No integration required. No PHI. You see HPDM's findings before committing to anything further.
Tier One
Strategic Advisory
$25,000 – $75,000 · SOW-Based
Market prioritization. Data access structuring with health plan and IDN clients. Performance participation design. Paid engagement with defined deliverables and a documented path to Tier Two.
Tier Two
Data Validation
$50,000 – $150,000 · Fixed Fee
HPDM deployed against your actual population. Full clinical data stack: EHR, Claims, Encounter, ADT, HIE feeds. Validated gap analysis with contract-ready figures. Secure analytics container. Your data. Your numbers.
Tier Three & Four
Live Intelligence & Full Activation
Platform License + Revenue Share
Ongoing HPDM scoring, 510 KPI reporting, continuous gap monitoring, and full six-layer activation through to All-Payer Shared Savings. We take risk alongside you on the outcomes we identify.

The Architecture Problem.
Read the full argument.

The Architecture Problem: Thirty Years of Clinical Evidence That Healthcare Data Was Never Built for the Person — and the Architecture That Finally Is
David D. Nethaway, MBA, CSM · Hypersphere Health, Inc. · May 2026
"Healthcare's most persistent failures are not mysteries... The clinicians seeing them have been right all along. The data systems built to support them have not been. Not because the technology failed — but because the architecture underneath every tool, every platform, and every analytics solution deployed in the last thirty years was built around a billing code, not a person."
Status: Working Paper
Pages: ~14
Citations: 13 peer-reviewed sources
Published: SSRN Preprint (pending)

Six sections. One argument.

01
What Clinicians Have Been Saying
IOM 2001–2012, AHRQ cost concentration, preventable admissions, SDOH invisibility
02
Why the Data Never Caught Up
HCC design intent, billing code as organizing unit, comorbidity interaction problem
03
The Cost of the Gap
OIG MA payment accuracy, $750B IOM waste estimate, RAF delta documentation
04
What the Architecture Needs to Be
Six logical requirements derived from the evidence. Why incremental improvement fails.
05
The HPDM Architecture
Four shells, three axes, Z-axis interaction engine, patent-pending disclosure
06
Conclusion
The architectural answer to thirty years of documented gaps — now patent-pending.
Request the Full Paper

David D. Nethaway
MBA, CSM · Principal
Hypersphere Health, Inc.
Naya Advisory Services, Inc.
Exercise Physiologist / Former EMT-Specialist
US Army Combat Lifesaver
5,000+ hours direct clinical contact
MBA Healthcare Management, U. Phoenix
Pursuing PgMP and TOGAF 10.0

Before HPDM, David Nethaway spent twenty years inside healthcare operations — as a clinician, as a program manager, and as a strategic advisor to health plans and government programs. He has seen the data architecture problem from every angle.

Hypersphere Health was founded on a single premise: the architecture was wrong, and fixing it required starting over — with the person at the center.

Naya Advisory Services provides the consulting engagement vehicle. Hypersphere Health, Inc. is the platform company. The patent-pending HPDM architecture is the foundation of both.


If this reflects what you're seeing
in your data — let's talk.

Not to sell you something. To understand whether HPDM applies to your specific problem. Twenty minutes. No deck. No pitch.

Tier Zero assessments are available immediately · Fixed fee · No integration required · No PHI