Skip to content

Science

Four analytic products, and what each one may not be used for

These products are deliberately not interchangeable. Most harm in this class of system comes from reading one product as another, so each contract states its permitted and non-permitted interpretations.

Definitions

Product contracts

Phenotype detection — p_osa(x, t)

Permitted and non-permitted interpretations for Phenotype detection — p_osa(x, t).
AspectStatement
QuestionHow likely is it that an unrecognised condition is present now, given the governed record at time t?
PermittedPrioritising evaluation capacity; describing detection likelihood at population level
Not permittedDiagnosis; treatment initiation; denial of coverage; reporting as an event risk
Refuses whenRequired features are UNKNOWN or CONTRADICTED, or the artifact is not permitted in this environment

Event prognosis — r_18(x, t)

Permitted and non-permitted interpretations for Event prognosis — r_18(x, t).
AspectStatement
QuestionWhat is the risk of a defined severe event within 18 months for a population with these characteristics?
PermittedPopulation stratification; capacity planning; monitoring design
Not permittedIndividual determinism; automated denial; substituting for diagnosis or for phenotype likelihood
Refuses whenObservation history is insufficient, or temporal leakage protection fails

Conditional intervention benefit

Permitted and non-permitted interpretations for Conditional intervention benefit.
AspectStatement
QuestionFor this indication and protocol, what benefit would this intervention be expected to produce?
PermittedComparing protocol-defined strategies under stated assumptions
Not permittedA universal drug score; cross-indication transfer; unqualified benefit claims
Refuses whenNo validated indication artifact exists — the current state in this build

External clinical/control arm (ECA)

Permitted and non-permitted interpretations for External clinical/control arm (ECA).
AspectStatement
QuestionCompared with a fit-for-purpose external comparator, what would outcomes have been under the alternative strategy for a protocol-defined population?
PermittedTarget-trial emulation with a frozen specification and prespecified analysis plan
Not permittedGenerating a member alert; acting as a risk score, spend calculation or generic comparison dashboard
Refuses whenOverlap is inadequate, positivity is not established, or the comparator is not fit for purpose

Use cases

Worked examples

These are configured demonstrations over synthetic data. None of them reports a validated production result.

Configured examples

Configured use cases and their explicit limitations.
Use caseWhat is configuredLimitation
OSA-18Direct-mining priority surface pairing detection likelihood against 18-month event riskIllustrative synthetic surface only; no validated coefficients, no clinical finding
Cognitive / FINGER-informedMultidomain trajectory framing with an endpoint selector and OSA-burden evidenceDomain framing only; no validated trajectory model and no cognitive diagnosis
KeytrudaIndication-specific protocol builder with comparator and estimand supportFirst configured example; refuses in the absence of a validated indication artifact

No model in this environment is validated. Experimental models never auto-deny therapy, ECA never produces a member alert, and a missing value is never treated as a negative finding.

Technical note: ECA-Agent. ECA-Agent is the RISX Science method and packaging standard by which a validated analytic use case becomes a bespoke, governed analytic agent inside a client environment: a locked, versioned artifact combining a diagnostic model, a prognostic model, a conditional causal component where justified, and an economic decision policy, with monitoring and controlled retraining. It is not a chatbot and not an autonomous clinical decision-maker. ECA-Agent: Analytic Methodology and Deployment Lifecycle is restricted; authorized technical, reviewer and administrator accounts may sign in to read it.