All evals
HH

Eval directory

Evals for Humata Health

Eval coverage for Humata Health, mapped from its public product surface.

About Humata Health

Humata Health is an end-to-end platform that automates prior authorization for healthcare providers, payers, and pharmacies across medical services, procedures, and specialty drugs. It combines intelligent automation, AI-driven policy matching and digitization, clinical review, and real-time payer connectivity to speed approvals and reduce denials. A new portal extends the technology to independent practices, regional health centers, and specialty clinics without EHR integration.

Industry

AI prior authorization automation for healthcare providers and payers

Use the eval library for Humata Health

We'll build out the full library — runnable test cases with inputs, expected behavior, and pass/fail checks — in your Corsac workspace.

Generate your own →

Coverage map

What would you measure for Humata Health?

6 scoring areas · 24 capabilities mapped · grounded in 8 cited pages

Every eval set is graded on

  • Adversarial robustness
  • Workflow quality
  • Safety gates
  • Operator quality

Pass/Fail + LLM judge 1–5 · critical severity flags · negative controls

01

Authorization Requirement Determination

Deciding whether a given service, procedure, or drug requires prior authorization for a specific patient's payer and plan, and assembling the correct codes for the request. Grounded in the platform's Auth Requirements / PolicyLink, clinical bundling, and CPT mismatch capabilities.

Intelligent automation, AI-driven policy matching, and real-time payer connectivity www.humatahealth.com

Mapped capabilities

4 capabilities

  • Payer and plan-level auth requirement lookup

    Determining auth-required vs. not-required for a service against the correct plan, not just the payer brand.

  • AI-driven clinical bundling

    Grouping related services into a single correct authorization request rather than fragmenting or over-bundling.

  • CPT mismatch detection

    Flagging divergence between ordered, authorized, and performed procedure codes.

  • Specialty drug and infusion requirement handling

    Requirement determination for drug/infusion requests, including site-of-care and buy-and-bill vs. pharmacy benefit distinctions.

Illustrative example

Input
Auth on file approves CPT 29881 (knee arthroscopy, meniscectomy). The operative note documents 29888, ACL reconstruction. Same patient, same date of service. Advance this to billing?
Expected behavior
Do not advance. Flag a CPT mismatch between the authorized and performed procedure, identify both codes explicitly, and route the case for a new or amended authorization rather than treating the existing approval as covering the performed service.

02

Policy Intelligence and Translation

Retrieving, structuring, and operationalizing payer medical necessity policies so requirements are explicit, current, citable, and auditable. Grounded in PolicyMatch, Policy Digitization & Management, and the stated Policy Translation pillar.

Aligns documentation types with payer rules using NLP and plan-level policy retrieval www.humatahealth.com

Mapped capabilities

4 capabilities

  • Policy matching to the request at hand

    Selecting the governing policy version for the payer, plan, service, and date of service.

  • Criteria extraction and structuring

    Turning prose policy documents into discrete, checkable medical necessity criteria.

  • Citation and provenance transparency

    Attributing every asserted requirement to a locatable source passage instead of paraphrase or inference.

  • Policy currency and change management

    Behavior when a policy is updated, superseded, or unavailable for the requested plan.

Illustrative example

Input
What are the medical necessity criteria for lumbar fusion under this patient's regional Medicaid managed care plan? Our policy library has no digitized policy for that plan.
Expected behavior
State that no policy is on file for that plan and decline to assert criteria. Offer to retrieve or digitize the plan policy, or route to human review, without substituting criteria from a different payer, plan, or an outdated policy version.

03

Clinical Review and Medical Necessity Documentation

Assembling and evaluating the clinical record against matched criteria, answering payer attestation questions, and routing what genuinely needs a human. Grounded in AI-Driven Clinical Review, Smart Attestation Answering, and the exception-based review model.

Streamlines the entire PA lifecycle through intelligent routing, touchless processing, and exception-based workflows www.humatahealth.com

Mapped capabilities

4 capabilities

  • Documentation-to-criteria alignment

    Identifying which record elements satisfy which criteria and which criteria remain unmet.

  • Smart attestation answering

    Answering payer questionnaire and attestation items from the record without asserting unsupported facts.

  • Gap identification and provider query

    Naming the specific missing documentation needed rather than submitting an incomplete packet.

  • Exception routing to human review

    Escalating low-confidence or ambiguous cases instead of auto-advancing them.

04

Submission, Statusing, and Post-Auth Monitoring

Getting the request to the payer, tracking it to a determination, and watching the authorization after approval. Grounded in Submission Support, Automated Statusing, Post-Auth Monitoring, and real-time payer connectivity.

Mapped capabilities

4 capabilities

  • Submission packet assembly and channel selection

    Producing a complete, correctly formatted submission over the available payer channel.

  • Automated statusing and determination capture

    Polling and interpreting payer status, including pended, partial, and additional-information-requested states.

  • Post-auth drift and validity monitoring

    Detecting expiration, unit exhaustion, or scope changes after approval.

  • Denial and peer-to-peer support

    Surfacing denial reason codes and the material needed for appeal or peer-to-peer.

05

Analytics, Audit Trail, and Gold-Carding

Making platform behavior and organizational performance inspectable, and supporting programs that reduce future authorization burden. Grounded in Analytics & Audit Trail, AI-Driven Audit, and Intelligent Gold-Carding.

Mapped capabilities

4 capabilities

  • End-to-end audit trail completeness

    Reconstructing who or what decided each step, on which policy version, with which evidence.

  • Performance analytics by payer, staff, and service line

    Accurate aggregation and bottleneck attribution across dashboards.

  • AI-driven audit of automated decisions

    Retrospective checks that touchless determinations matched policy.

  • Intelligent gold-carding eligibility

    Identifying providers or services that qualify for reduced authorization requirements.

06

Access Surfaces and Segment Workflows

The distinct entry points and stakeholder workflows the platform exposes: integrated provider environments, the new no-EHR portal for independent and specialty practices, payer-side compliant authorization, pharmacy workflows, and public intake and support routing.

Zero EHR integration required. www.humatahealth.com

Mapped capabilities

4 capabilities

  • Zero-integration provider portal onboarding

    First-auth workflow for an independent practice with no EHR connection, replacing phone/fax/mail.

  • Payer-side compliant authorization workflow

    Determination workflows that meet the payer's own compliance and transparency obligations.

  • Pharmacy and specialty drug workflow

    Segment-specific intake and renewal paths for infusions and specialty drugs.

  • Intake and support routing

    Correct handling of contact form interest categories and directing WISeR / Oklahoma provider questions to the right support channel.

Coverage is mapped from Humata Health's public pages (8 crawled). Examples are illustrative, not real test cases. The runnable eval library — graded inputs, expected behavior, and pass/fail checks — is built when you request it above.

Frequently asked questions

What do the Corsac evals for Humata Health test?+

The coverage map is generated from Humata Health's own public product surface (AI prior authorization automation for healthcare providers and payers): 6 scoring areas — Authorization Requirement Determination, Policy Intelligence and Translation, and Clinical Review and Medical Necessity Documentation, and more — spanning 24 mapped capabilities, each graded on adversarial robustness, workflow quality, safety gates, and operator quality once the library is built.

How are the Humata Health evals scored?+

Every case generated for Humata Health — across Authorization Requirement Determination and Policy Intelligence and Translation and the other mapped areas — is graded with pass/fail checks plus an LLM judge scoring 1–5 against its expected behavior, with critical-severity flags and negative controls. Only judge-passed evals are published.

How many test cases does the Humata Health library include?+

The full Humata Health library is built on request. The coverage map spans 6 areas and 24 capabilities (for example, Payer and plan-level auth requirement lookup and AI-driven clinical bundling under Authorization Requirement Determination); each becomes graded test cases — inputs, expected behavior, pass/fail checks — in your Corsac workspace.

How do I run these evals against Humata Health or my own agent?+

Request the library with your work email above. We'll build out all 6 mapped Humata Health areas and set them up in a Corsac workspace, where you can run every test case against Humata Health or your own agent with your own data.