Working in collaboration with Mayo Clinic
Patent pending · USPTO Track One
Healthcare governs what it can measure.
Behavioral load has no governed reference number.
We built the candidate — and the category it anchors.
Stress Number™ quantifies behavioral load across Home, Work and Social. Clinically validated in research conducted at Mayo Clinic, peer-reviewed and published in 2018.
Deployment-agnostic behavioral health technology designed to reach 100% of a population — not only the share that shows up.
Peer-reviewed clinical validation
Mayo Clinic study CLS-20308379 · N = 292
Patent Pending · USPTO Track One
A 20-minute institutional overview · executive and actuarial scoping available · all materials under NDA
AI has abundant signal. It has no validated behavioral referent. Where this sits relative to AI →An upstream measurement and routing layer that sits ahead of existing behavioral-health, coaching, navigation and care-management assets. Population measurement without an encounter, and the repeat administration a quality measure requires. The benefit: more members measured means more members identified and routed before the cost appears in claims.
How it deploys →HRV, sleep and recovery provide physiological signals associated with stress. Stress Number™ adds a direct self-reported measure of behavioral load across three settings, with peer-reviewed clinical validation behind the instrument.
How it deploys →Two points of entry, both running on the same standard. The Signal is where a member takes the Stress Number™ — a 30-question assessment producing a personal score across Home, Work and Social, with routing through the Smart Referral Engine™. The Visibility is Population Scale, the licensee dashboard.
Population Scale is where the measure becomes infrastructure. Individual responses aggregate — de-identified — into a leading-indicator view that rolls up and filters across an entire book of business: by company, by geography, by department, by site. An administrator sees where behavioral load is concentrated in a covered population without seeing any individual in it. That aggregation is what makes the measure usable for quality reporting, actuarial modeling and capacity planning rather than for one person at a time.
Domain distribution and movement show where behavioral load is concentrated — Home, Work or Social — creating a routing signal that a single undifferentiated score cannot provide.
Filter and visualize as heat maps. Configure Smart Referral Engine™ routing rules at the item, domain and score level. Trends and benchmarks convert repeat administration into a pre/post measurement record.
Population figures use the 0–100 normalized display convention — not the Stress Number™ composite scale, and not a clinical decision threshold.
This is the difference between a questionnaire and infrastructure. Measurement that ends at a number moves the burden to a care team. Measurement that routes moves the person.
Thirty questions, any channel a population already uses. No appointment, no clinician time.
A composite and three domain scores — Home, Work, Social — returned immediately.
Domain resolution shows where the load sits. Social isolation is a different problem than work strain, and routes differently.
Smart Referral Engine™. Rules configured at item, domain and score level match the person to support that is actually available to them locally.
The administrator sees population distribution, heat maps and routing utilization — without seeing individuals.
Repeat administration on the same population produces the before and after a quality measure requires.
Stress Number™ measures behavioral load across three settings — Home, Work and Social — through thirty questions completed in minutes. Not a symptom checklist. A measure of behavioral load across the environments where it accumulates.
Existing instruments ask what a person is experiencing now, and they score symptoms once those symptoms have appeared. Stress Number™ measures behavioral load across the settings where it accumulates — upstream of the symptom, not alongside it.
It does not require an appointment, a clinician, or a diagnosis to be administered. It runs wherever an organization already reaches its members, which is what makes measuring an entire covered population possible rather than only the share already in care.
Thirty questions, three named domains, a documented scoring specification and a version record. A person sees where their load sits. An administrator sees where a population's load sits. The methodology is citable.
On scales and thresholds. The 0–100 values shown are a normalized display convention for domain reporting. They are not the Stress Number™ composite scale, and the colored bands are a presentation convention rather than clinical decision thresholds. A candidate composite threshold has been derived internally from the retained research dataset; it is not what the 2018 publication established, it has not been independently reproduced, and it is not presented here as a validated or clinically adopted cutoff. The 2018 publication established correlational clinical validity against two established comparators — not a decision threshold.
Every system already pays for it — in claims, in absence, in turnover, in treaty losses, in quality scores. What none of them has is a number for it before it becomes a diagnosis.
Behavioral risk is identified once symptoms cross a diagnostic threshold. By then the load has usually accumulated for years, and the intervention is treatment rather than prevention.
Screening is encounter-bound and episodic. It reaches people already in care, when screening is indicated. Most covered populations are never reached by validated behavioral measurement at population scale.
The instruments in routine use ask what a person is experiencing now. They do not resolve behavioral load across Home, Work and Social — the environments in which that load accumulates and may remain addressable before diagnosis.
Demonstrating improvement requires measuring the same population twice — which favors measurement that can reach people beyond the clinical encounter. Penetration methodology and estimates: Pre-Diagnostic Index™ → · Read the category →
The case for an upstream measure is not that it would be useful. It is that the absence of one is already being paid for — in reserves set on inference, in benefits spend allocated without a target, in quality scores that move on something no one can see. These are the positions where that cost lands first.
Behavioral claims are rising against reserves set without an upstream predictor. The variable driving the trend is the one variable not measured before the claim.
Behavioral factors influence claim duration and return-to-work, but enter the file only after injury. There is no pre-loss measure to underwrite against.
Improving or Maintaining Mental Health moves to weight 3 for 2027. Plans must demonstrate movement on a population they cannot currently measure before it presents.
Behavioral spend is allocated by engagement guess rather than measured need. The CFO cannot show what the spend bought, because there was no baseline.
Low utilization is the category’s structural problem, and it persists because outreach is broadcast. Without a population measure there is no pre and no post.
Workforce burnout drives turnover, safety events and experience scores. The exposure is managed with survey instruments that reach a fraction of staff, episodically.
Treaty pricing carries behavioral risk as an inferred variable inside a blended trend assumption. It is priced, but it is not measured.
These systems hold the employment record, the absence record and the benefits enrollment, and no governed behavioral variable to sit against any of them.
The absence is already being paid for. It is simply paid at different points in the chain, by different holders, in different currencies — which is why no single seat has ever been able to see the whole exposure.
We measure behavioral load upstream.
We translate it into actuarial exposure.
We route people to the right support.
One measure resolved across three domains — Home, Work and Social. Thirty items, minutes to complete, portal-agnostic.
The instrument →The layer that converts the coordinate into actuarial exposure. Derived from federal and peer-reviewed published sources.
The economics →Domain scores indicate what kind of support may fit; the referral layer surfaces what is actually available locally.
The routing →The operating system that holds them — the scoring architecture, the governance, and the record of who measured what, when, and under which version.
The standard →O2OS™ is portal-agnostic. It runs wherever an organization already meets its members — a member portal, a benefits platform, an EHR patient gateway, a secure link, or SMS. It does not replace an EHR, compete with one, or require a clinical encounter to function.
This is what the operating-system architecture buys an institution. Capabilities are activated, not installed. Measurement, routing, EHR integration, reimbursement support and governance are separable — an organization takes the ones it needs and leaves the rest, without displacing the systems it already runs on.
A browser link. Nothing to install, no app store, no device management, no IT provisioning. Kiosk, tablet, phone or staff-assisted entry all work. Where integration is configured, the same interaction can return structured data to institutional systems, including the health record.
Sits alongside existing systems rather than inside them. Epic integration paths include SmartForms, flowsheets, launch buttons and FHIR/HL7 bundles, with FHIR export available as an administrator capability. EHR integration is optional — the system runs without it.
Three domain outputs per person, with the governed composite scoring architecture available under protocol. Routing recommendations from the Smart Referral Engine™. An administrator dashboard with population composite, domain distribution and trend. Pre/post on repeat administration.
No clinician time required for administration. No replacement clinical workflow. No credentialing requirement for assessment administration. The measurement architecture is licensed and can be deployed within existing enterprise infrastructure rather than requiring a replacement platform.
A structured briefing under NDA. The full evidence state, the scoring architecture, and the deployment model. Typically one session.
A defined population, a defined window. Completion rates, domain distribution, routing utilization and workflow burden — measured in your environment, not asserted in ours.
Co-designed deployment where the use case needs it. Integration, workflow and reporting built to your environment rather than around a fixed product.
Population-scale deployment under license. Covered-life economics, defined term, and a sector position where one is available.
A pilot does not require a settled decision threshold. It answers whether the instrument reaches your population, what it finds, and what it costs to run — questions the published validation cannot answer for your environment. Start a conversation →
Models in this domain are trained on claims, clinical notes and physiological telemetry. All three are downstream of presentation. A wearable can observe that heart-rate variability declined, that sleep fragmented, that recovery fell. It cannot say what changed, in which part of a person’s life, or whether it matters — because there is no clinically validated behavioral measure for those signals to be calibrated against.
Thirty structured items, three named domains, a documented scoring specification and a version record — validated against two established clinical comparators. This is the referent physiological telemetry currently lacks.
HRV, sleep and activity are correlates of stress without a clinical anchor. A self-reported measure administered alongside them gives the passive layer something to be validated against.
The Smart Referral Engine℠ operates on configured rules today. With population-scale measurement and tracked referral outcomes, the same architecture supports routing that improves against observed results.
An instrument administered after someone presents clinically produces a dataset with selection already built into it — every subject had already reached the system. Measurement deployed upstream across a whole population observes people before the event exists. Where outcomes are later linked under appropriate governance, that is the structure longitudinal prediction requires.
A buyer should know precisely which claims are established and which are still in development. The distinction is stated here rather than discovered in diligence. See the standard →
Independent institutional reproduction represents the next scientific layer. The published validation, retained participant-level dataset and current operating architecture stand independently of that workstream.
The federal payment system is converging on a single structural need at the same moment. Behavioral health screening remains largely encounter-bound and episodic, so most covered populations are never reached by validated behavioral measurement at population scale. Demonstrating improvement requires measuring the same population twice.
CMS-0057-F · payer API requirements phasing in from 2026
CMS ACCESS Model · outcome-aligned reimbursement for behavioral health
Star Ratings · Improving or Maintaining Mental Health moves weight 1 → weight 3
Consolidated Appropriations Act · plan sponsors must demonstrate vetted behavioral spend
The obvious question about a 2008 filing is why it has not already sold. The answer is that infrastructure cannot be sold into a system that carries no obligation to use it. For most of those eighteen years there was no federal requirement to measure behavioral load upstream, no reimbursement pathway attached to it, and no quality measure that moved on it. Organizations were not declining to buy an upstream behavioral measure. There was no position in the system for one to occupy.
A billing code is permission, not obligation. CPT 96127 has existed since 2015 and reimburses a few dollars per instrument — it allows a brief behavioral assessment to be billed, but requires no one to measure anything, and creates no accountability for the result. The clocks above are different in kind: they attach payment, quality scoring and fiduciary duty to behavioral outcomes. That is what converts a capability into a requirement.
The architecture was built before the requirement existed, which is why it is available now rather than under construction.
Eight Star measures carry a weight of 3. Behavioral load may sit upstream of all of them — one directly, three supported by published literature, four hypothesized and untested. See the exposure map →
The peer set — FICO, Bloomberg, MSCI, S&P Global, HL7 — achieved position through progressive satisfaction of these criteria, not through a single designation event. Stress Number™ is the reference instrument at the core of the O2OS™ architecture. Each criterion below is stated with its status and the documentary basis behind it.
Nine of eleven criteria supported by documentary record. Two actively in motion. Chain of custody: Mayo Clinic research → Stress Number™ → O2OS™ reference-standard architecture. Read the standard →
Six federal mechanisms. Each a real designation, publicly documented, and independently defensible — not dependence on a single program that could be modified.
The closed circuit. CMS-0057-F creates the pipe. The 2027 Star Ratings provide the check. O2OS™ is architected to sit between the two — using the pipe to deliver the leading indicators that secure the check. Also aligned: CPT 96127 / 96138, Medicaid 1115 waivers, and the CMS ACCESS Model.
Access is governed by a structured tier system, published rather than pitched — the way standards bodies publish membership and licensing criteria. The measurement architecture is licensed and can be deployed within existing enterprise infrastructure rather than requiring a replacement platform.
Access and licensing are structured to preserve the independence required for multi-institutional adoption and future reference-standard recognition. Read the standard →
An instrument, a filing, a clinical study, and a retained dataset that still reproduces its own published statistics. Precision held over time is the asset.
About
the corporation →
Products require adoption. Standards require recognition.
O2OS™ contains the foundational assets required to pursue clinical measurement reference-standard recognition — formal recognition remains ahead.
A peer-reviewed clinical-validation asset, a working deployment architecture, retained research data, and a patent-pending position. The asset is available for licensing, paid pilots, design partnerships and enterprise deployment today. Independent reproduction and reference-standard recognition constitute the next institutional layer.
Institutional and enterprise inquiries: contact@theoxygenplan.com · Eric G. Lucas, Founder & Chief Executive Officer · Direct peer-to-peer contact welcomed.