Necessity is a chart-read
Every UM and post-payment review is a re-read of records the system has already touched. Owl reads every page of the chart, walks MCG, InterQual and LCD criteria element by element, and tells you which bucket each line lands in (approve, deny, pend or refer), cited for member, provider and IRO.
KL grade 3 imaging, BMI 31.4, six months conservative care including NSAID + 14 PT visits + two IA-CS injections, surgeon in-network. Every element documented in the chart.
Inpatient site supported by anesthesia history (severe OSA, AHI 38) and prior post-op respiratory event documented in the discharge summary from 2023.
No CMS step-therapy required for this CPT under our medical policy. Pharmacy step requirements do not apply to surgical authorization.
CPT 27447 is FDA-cleared and on the plan’s standard surgical inventory. No experimental override needed.
Auto-approve
Drafted with MCG and EOC cites
11s, 1.4s
Full-chart read, answer time
$0
Cost of pending an approvable case
Three AI agents on a UM file. Reading. Cross-referencing. Answering.
One PA: 312 pages, 11 seconds, four questions answered.
Anonymized composite. Member 64F, CPT 27447 (right TKA), submitted by an in-network ortho group with full chart attached.
External data only. None of this lives in your CMS. Owl gathered it from public records, professional networks, registries and the open web, then resolved it to this claimant.
MCG 27th ed. and InterQual SmartSheet for TKA both mark every required element met. Plan-cohort auto-approve rate on this CPT + member profile is 94%. Auto-approval is in-band for plan policy and CMS standards.
Owl’s IRO database (n=180K independent-review decisions across 14 states) shows denial of TKA when MCG criteria are met is overturned at 71%. Denial would create CMS Star measure exposure and provider abrasion without medical-necessity basis.
CMS Care Compare + Hospital Compare show surgeon’s 30-day readmit at 11.4% (specialty peer 4.2%). Not denial-relevant; route to case-management for post-op outreach to mitigate readmit risk.
Datavant + Epic Care Everywhere pull surfaces a prior MI from 2022 not in the submitted chart. No cardiac clearance is documented for this surgery. Recommend pend for clearance, then approve: small delay, large risk reduction.
Yes, approvable. The member meets every MCG TKA criterion: KL grade 3 imaging, BMI under 35, six months of failed conservative care including NSAID, 14 PT visits and two IA-CS injections. Surgeon is in-network. The single open item is documented cardiac clearance given the prior MI surfaced from external EHR. Recommend approve with a note to the surgeon to obtain clearance before scheduling. Determination language with citations to MCG and the member EOC is drafted below.
Substantial. Owl’s IRO corpus shows 71% overturn for denials on this fact pattern. Beyond the financial exposure, denial would harm the plan’s CMS Stars appeals-timeliness measure and create avoidable provider and member abrasion. There is no documented medical-necessity basis to deny.
What the world says about your Medical claims review claimants, none of which lives in your CMS.
Internal claim data is half the picture: it tells you what the claimant told you. OwlSignal continuously gathers external data (public records, professional and corporate registries, court dockets, social and open web) and matches it back to the claimant. These are the families that move the needle on Medical claims review.
MCG, InterQual, Hayes, ECRI, NCCN, ACR Appropriateness: the criteria libraries that determine medical necessity. Walked element-by-element against the chart, with the version and edition cited on every decision.
Independent Review Organization (IRO) decisions from state DOIs and CMS: what gets overturned, what survives. Owl’s corpus is 180K decisions across 14 states, refreshed monthly.
CMS Care Compare, Hospital Compare, Physician Compare, OIG LEIE, state license sanctions, Owl’s cross-payer utilization graph, to surface provider context the chart alone won’t reveal.
Longitudinal medical history from outside your data, where the law and the member’s consent allow. Surfaces prior conditions, prior procedures, and prior failed therapies the submitted chart doesn’t include.
EOCs, medical policies, LCDs, NCDs, drug formularies and benefit grids. Read once, indexed forever, and re-walked on every determination so policy changes propagate immediately.
OIG annual Workplan, CERT national error-rate data and HHS PERM samples: the patterns that determine pay-and-chase exposure and FWA referral defensibility.
Numbers from health plans running Owl on medical claims review in production.
Built for the regulatory shape of Medical claims review.
Lives where your Medical claims review files live.
Core admin
HealthEdge HealthRules, Cognizant Trizetto Facets, QNXT
UM platforms
GuidingCare, Aerial, MedHOK, ZeOmega Jiva
EHR / records
Epic, Cerner, Allscripts, MRO, Verisma, Datavant
Criteria libraries
MCG, InterQual, Hayes, ECRI, NCCN, ACR
Document intake
Box, OnBase, Documentum, ImageRight, faxes via SR-Fax
Identity & SSO
Okta, Azure AD, PingFederate, SAML, SCIM
Bring us a week of UM reviews. We’ll show you the bucketing on day one.
Two-week pilot. Your charts, your tenant, your medical policy. We measure cycle time, auto-approval rate, IRO survival rate and reviewer hours saved against your own ground truth.