Buyer's Guide

Best risk adjustment software for IPAs and ACOs in 2026.

~9 min readBy Pelica Health teamPublished May 27, 2026Updated August 13, 2026Reviewed by Lalit Kundu

TL;DR

The best risk adjustment software for an IPA or ACO in 2026 does both prospective and retrospective capture in one place, applies V28-aware trumping logic, keeps a RADV-defensible chain-of-custody on every HCC, and shows the full submission lifecycle from 277CA through MOR. Point coding engines do parts of this well. The differentiator is whether one record is shared across every team that touches the member.

Vendor comparison

The vendors below are all credible in risk adjustment. They differ in where they focus: some at the point of care, some in retrospective chart review, some as broad data platforms. Categories are descriptive, not pejorative.

Comparison reflects each vendor's publicly stated positioning as of August 2026. Capabilities vary by contract and scope; confirm current details directly with each vendor.
Vendor	Focus (prospective / retrospective / both)	AI approach	RADV / audit support	Submission-lifecycle visibility	Best-fit org
Navina	Prospective, point-of-care	Clinician copilot; EHR-native condition detection	Evidence linked to the chart	Not its primary focus	Physician groups and ACOs working in the exam room
Reveleer	Both; retrospective abstraction plus prospective suspecting since EVE Hybrid AI (January 2026)	EVE for chart review; EVE Hybrid AI pairs generative extraction with clinician-authored deterministic rules	All-in-one RADV Audit SaaS (October 2025); manages CMS and RADV-IVA submissions	Submission and audit-focused	Plans running scaled chart review and audit
Apixio (Datavant)	Both; retrospective, prospective, concurrent	NLP/ML coding predictions plus coder review	HCC Auditor; compliance solutions	Coding and compliance-focused	MA and ACA organizations scaling coding accuracy
RAAPID	Both; retrospective and concurrent	Neuro-symbolic AI plus knowledge-graph NLP	MEAT-based audit trail; RADV solution	Coding and audit-focused	Plans and groups wanting defensible coding at scale
ForeSee Medical	Prospective, point-of-care	NLP disease-detection engine in the EHR	Source-document hyperlinking (InstaVu)	Not its primary focus	Practices and plans wanting EHR-embedded suspecting
Innovaccer	Both, via the Galaxy payer platform and Gravity agents	Unified data model plus an agent layer; 2026 Best in KLAS for payer data analytics at 90.5	Part of broader risk and quality workflows	Within the platform	Enterprise plans standardizing on one data platform
Arcadia	Retrospective-led, analytics	Data lakehouse and analytics	Risk analytics, not a coding engine	Analytics view	Systems and ACOs wanting a data foundation
Datavant	Retrospective coding and retrieval at scale, plus outsourced coding services	Clinical Insights Platform; NLP coding with certified coder review	2026 Best in KLAS in Risk Adjustment (Coding, Retrieval and Compliance) at 90.5	Coding and retrieval-focused	Plans outsourcing retrieval and coding volume
Cozeva (Vatica Health)	Prospective, point-of-care and in-home	PayerOne Risk with native-FHIR certified measure engines	2026 Best in KLAS in Risk Adjustment: Point of Care and In-Home at 91.2	Not its primary focus	Plans and groups prioritizing point-of-care and in-home capture
Pelica Risk Adjustment Copilot	Both, real-time, on one shared record	Copilot plus action layer that executes the follow-through	Chain-of-custody per HCC; trumping applied live	277CA through MOR, per diagnosis, in one view	Risk-bearing IPAs, ACOs, and plans replacing vendor sprawl

Why risk adjustment software is a 2026 problem, not a 2024 one

The CMS-HCC V28 model is fully effective for payment year 2026, completing a three-year phase-in. V28 cut the ICD-10-to-HCC crosswalk from roughly 9,797 valid codes under V24 to about 7,770, and CMS's actuaries projected an average MA risk score reduction of about 3.12%. Diagnoses that triggered an HCC under V24 may no longer do so. Read our V28 readiness playbook for the full picture. If you also carry commercial marketplace membership, confirm the tool maps to HHS-HCC, the separate ACA risk model, and not to CMS-HCC alone.

The operational consequence is simple. If your software only works after claims settle, it will tell you what you lost after the window to fix it has closed. The capture decision now has to happen at or before the encounter. That single shift reorders what "best" means in this category.

One 2026 development belongs here even though it is a Stars rule. The CY2027 Medicare Advantage and Part D final rule, published April 6, 2026, removes 11 measures from Star Ratings calculations beginning with the 2027 measurement period. It does not change the HCC model, and V28 remains the binding constraint on capture. It does change where the same budget goes, because risk and quality budgets are set in the same room and a shorter measure set means each remaining Stars lever is worth more.

What to look for in risk adjustment software

Prospective and retrospective in one place

Prospective capture happens at or before the visit, so the diagnosis is documented within the year it is clinically valid. Retrospective review catches what was missed and supports recovery and audit. Running them as two disconnected tools means the two views never reconcile against the same member record. The best software does both and reconciles them.

V28-aware trumping logic

V28 expanded to 115 condition categories and tightened constrained groups, where related conditions share a single coefficient. Documenting more conditions in a constrained group does not raise RAF: only the highest-weighted member counts. Software should apply this trumping logic live, so coders are not chasing diagnoses that will not pay, and so redundant capture does not create audit exposure.

RADV-defensible chain-of-custody

Every captured HCC should carry a retrievable evidence packet: the source clinical note with MEAT support, supporting labs or imaging, any provider attestation, and the coder review record. This should be a property of the system, not a packet assembled by hand the week an auditor calls. Our RADV-defensible HCC guide details the design choices that hold up under audit.

Full submission-lifecycle visibility

A diagnosis is not captured until CMS recognizes it. Good software tracks each one through the 277CA acknowledgement, the MAO-004 (which reports diagnosis-level risk eligibility, additions, and deletions), the MMR membership and payment report, and the MOR model output report. MAO-004 rejections are an early signal that documentation will not survive RADV. Seeing the whole chain in one place is the difference between catching a problem and discovering it in a payment shortfall.

Point-of-care, pre-claim flagging

The highest-leverage feature is surfacing the specific at-risk condition during the encounter, with the clinical criteria required. Generic prompts ("review chronic conditions") do not move capture. Specific prompts ("documented CKD Stage 3a in 2024; confirm staging today") do.

Navina

Navina is a clinician-first copilot for prospective capture. It summarizes patient data from the EHR, HIE, and claims, surfaces suspected conditions at the point of care, and supports one-click documentation inside the chart. For a physician group whose main lever is the visit, Navina is well designed. It raised a $55M Series C on March 25, 2025 led by Growth Equity at Goldman Sachs Alternatives, and its Best in KLAS win in Clinician Digital Workflow was 2025, not 2026; TransformativeMed took that category in 2026. (Source: Navina risk adjustment.)

Reveleer

Reveleer is built for high-volume retrospective chart review. Its Evidence Validation Engine retrieves and parses records and populates abstraction fields for review, and the company manages CMS and RADV-IVA submissions. Calling it retrospective-only is no longer accurate. It acquired Curation Health in October 2024 for prospective, point-of-care rules with Epic, Cerner, and Athenahealth integrations, shipped an all-in-one RADV Audit SaaS on October 15, 2025, and launched EVE Hybrid AI for prospective diagnosis suspecting on January 20, 2026, pairing generative extraction with clinician-authored deterministic rules. In April 2026 it introduced a Clinical Data Repository it describes as a unified member-level clinical record across risk, quality, RADV, and care management. For plans that need scaled, on-time chart review and audit submission, Reveleer is a strong fit, and it now competes on the prospective side too. It sells to plans, and delegated provider groups sit on the other side of that contract, which we take apart in Pelica vs Reveleer alongside a job-by-job list of Reveleer alternatives. (Sources: Reveleer retrospective risk adjustment; EVE Hybrid AI announcement.)

Apixio (part of Datavant)

Apixio's value-based care suite has been part of Datavant since September 2024, so evaluate it as Datavant's coding engine rather than a standalone vendor. The technology itself covers retrospective, prospective, and concurrent coding, combining NLP and machine-learning predictions with certified coder review, plus an AI-powered HCC auditing solution. For MA and ACA organizations focused on coding completeness and compliance across the full review cycle, it remains a credible engine inside the larger Datavant platform.

Datavant

Datavant belongs on more 2026 shortlists than it lands on. It won 2026 Best in KLAS in Risk Adjustment (Coding, Retrieval and Compliance Solutions) at 90.5 and in Outsourced Coding at 94.9 against a category average of 88.4, both announced February 4, 2026, and its release cites 4.5 million charts coded annually at 98% coding accuracy. Its products in this category are the Clinical Insights Platform and Datavant Connect. If your constraint is retrieval and coding throughput, this is the strongest externally validated option on the page. (Source: Datavant 2026 Best in KLAS.)

Cozeva, part of Vatica Health

Cozeva won 2026 Best in KLAS in Risk Adjustment: Point of Care and In-Home Health Assessments at 91.2, announced the same day. It is no longer a standalone company: Vatica Health and Cozeva merged on October 31, 2025. Its PayerOne Risk product sits at the point of care, which is the axis this page argues capture has to move to under V28. (Source: Cozeva 2026 Best in KLAS.)

RAAPID

RAAPID uses neuro-symbolic AI and knowledge-graph-infused clinical NLP to produce an evidence-backed, MEAT-based audit trail for every HCC. It supports both autonomous retrospective review and a RADV audit solution. For teams that prioritize defensible, traceable coding, RAAPID's design leans into audit transparency. (Source: RAAPID retrospective risk adjustment.)

ForeSee Medical

ForeSee Medical embeds NLP-driven disease detection directly in the EHR, presenting risk-adjustment suspects and HCC recommendations at the point of care, with InstaVu hyperlinking suspects back to the exact page of source documentation. For practices that want prospective suspecting inside their existing workflow, it fits. (Source: ForeSee Medical HCC coding.)

Innovaccer and Arcadia

Both are enterprise data platforms rather than dedicated coding engines. Innovaccer's Galaxy platform unifies payer risk and quality on one data model with an agent layer on top, and it took 2026 Best in KLAS in Data Analytics Platform for Payers at 90.5 against a category average of 87.2, announced February 4, 2026. On April 15, 2026 the company said it is moving from software-access fees to outcome-based, per-task pricing, citing roughly $20 per prior authorization against roughly $100 for the manual equivalent, alongside a $250M commitment over three years to its Gravity agent platform. Arcadia provides analytics and a longitudinal record across a large population; Nordic Capital acquired a majority stake on July 2, 2025. For organizations whose first need is a clean data foundation, either is reasonable, and risk adjustment is one workflow within a larger platform. (Sources: Innovaccer 2026 Best in KLAS; Arcadia platform.)

How to read the one-data-model argument

Innovaccer publishes its own risk adjustment comparison, and its central claim is that one data model across risk and quality means a plan does not pay for the same chart twice. That claim is correct, and duplicate retrieval is a real line item. Reveleer makes a version of the same argument with its Clinical Data Repository, citing plans retrieving the same records an average of 2.5 times a year.

Paying twice for a chart is one problem. Submitting a diagnosis and never learning it was rejected is a different one, and it shows up in a payment shortfall rather than an invoice. Ask any vendor to show a single diagnosis traced from 277CA through MAO-004, MMR, and MOR in one view. Then ask the four questions that separate a platform claim from an operating one: who performs the next action after a suspect is raised, how long it takes from suspect to first contact with the practice, which system the evidence is written back into, and whether the outreach queue is shared with quality, pharmacy, and care management or run per program. Cross-payer gap consolidation is the version of that last question a buyer can test in a demo.

Per-task pricing on prior authorization is a genuine move toward paying for work rather than access. It also says nothing about who works a suspect list.

How Pelica's Risk Adjustment Copilot differs

Most of the vendors above do one part of the job well: prospective suspecting, or retrospective review, or audit trails, or analytics. Pelica's Risk Adjustment Copilot is built to do them together, in real time, on a record that the rest of the organization shares. For a concrete walkthrough of how the copilot runs a single capture end to end, see the anatomy of an AI HCC gap closure.

Real-time, both directions. At-risk HCCs surface before the encounter, and retrospective review reconciles against the same record, so prospective and retrospective never drift apart.
V28 trumping applied live. Constrained-group trumping is computed as conditions are captured, so coders work the diagnoses that actually pay.
Chain-of-custody by default. Every captured HCC carries its source note, supporting evidence, attestation, and coder review record, ready for RADV.
Full lifecycle in one view. Each diagnosis is tracked from 277CA through MAO-004, MMR, and MOR, so rejections are caught early.
One record across all teams. The same canonical record feeds quality, pharmacy, network, and care management, so a member with three at-risk HCCs gets one well-prepared encounter, not three separate calls.

2 to 4 weeks

From kickoff to a live Risk Adjustment Copilot on your feeds

~200,000

Patients managed live on the platform at our flagship customer

~10 hrs/week

Saved per user across risk and the other teams

At our flagship customer, a physician-led IPA in New York running risk on roughly 200,000 patients, the Risk Adjustment Copilot reached 100% team adoption across the platform. Across Pelica deployments, customers have lifted RAF by roughly +0.4 in two quarters with no new headcount.

The point is not a better coding engine. It is the same record under risk, quality, pharmacy, network, and care management, so capture is a coordinated operating cadence, not a siloed coding project.

If you are evaluating that one-record approach across more than risk adjustment, our guide to the best value-based care software compares the platforms that unify risk, quality, pharmacy, and care management rather than stacking point tools.

Sources and further reading

Risk Adjustment Software Directory: a vendor-by-vendor reference we publish, with a fact sheet and dated sources for each of the ten platforms named here
CMS: 2024 (V28) Model Software / ICD-10 Mappings
CSSC Operations: Encounter Data and risk adjustment report FAQs (277CA, MAO-004)
Navina: AI-powered risk adjustment software
Reveleer: Retrospective risk adjustment
Datavant: 2026 Best in KLAS in risk adjustment coding, retrieval and compliance, and outsourced coding (February 4, 2026)
Cozeva: 2026 Best in KLAS in risk adjustment point of care and in-home assessments
Federal Register: CY2027 Medicare Advantage and Part D final rule (April 6, 2026)
Reveleer: EVE Hybrid AI for prospective diagnosis suspecting (January 20, 2026)
Innovaccer: 2026 Best in KLAS results

Frequently asked questions

Common questions IPAs and ACOs ask when choosing risk adjustment software in 2026.

What is the best risk adjustment software for an IPA?

There is no single best tool for every IPA. The right choice depends on whether you need prospective point-of-care capture, retrospective chart review, or both, and whether you want a standalone RA engine or a platform that also covers quality, pharmacy, network, and care management on one record. An IPA running a full risk-bearing contract usually benefits most from software that does both prospective and retrospective capture in one place and shares that record across every team, rather than a point coding engine bolted onto a stack of other vendors.

Should risk adjustment be prospective or retrospective?

Both. Prospective capture happens at or before the visit, so the diagnosis is documented and submitted within the year it is clinically valid, which matters most under V28's tighter mappings. Retrospective review catches what was missed and supports audit and recovery. The weakness of running them as separate tools is that the prospective and retrospective views never reconcile against the same record. Doing both in one system is what closes that gap.

What makes a capture program RADV-defensible?

A RADV-defensible program attaches a retrievable evidence packet to every captured HCC: the source clinical note with MEAT support, the relevant labs or imaging, any provider attestation, and the coder review record. Trumping logic must be applied correctly so only the highest-weighted condition in a constrained group hits the claim. Chain-of-custody should be a system property, not a manual workflow assembled after an auditor asks.

Does V28 change which software I need?

Yes. V28 cut the ICD-10-to-HCC crosswalk from roughly 9,797 codes to about 7,770 and applies fully in payment year 2026, so diagnoses that triggered an HCC under V24 may no longer do so. Software that only works retrospectively will surface losses after the window has closed. Under V28 you need V28-aware logic that flags at-risk conditions before the encounter, applies trumping correctly, and tracks every diagnosis through the submission lifecycle.

What submission reports should risk adjustment software track?

At minimum the 277CA acknowledgement, the MAO-004 (which reports diagnosis-level risk eligibility and rejections), the MMR membership and payment report, and the MOR model output report. MAO-004 rejections are an early signal that a diagnosis will not survive RADV. Software that shows the full lifecycle per diagnosis in one place lets a team catch and fix problems before they become payment shortfalls.

Which risk adjustment vendors won 2026 Best in KLAS?

KLAS split the risk adjustment categories in 2026 instead of handing them to one vendor. Datavant won Risk Adjustment (Coding, Retrieval and Compliance Solutions) at 90.5 and Outsourced Coding at 94.9 against a category average of 88.4. Cozeva, now part of Vatica Health, won Risk Adjustment: Point of Care and In-Home Health Assessments at 91.2. Innovaccer's Galaxy won Data Analytics Platform for Payers at 90.5 against a category average of 87.2. All three were announced February 4, 2026. The split is itself useful information: no single vendor owns the category, so buy by the job you need done, not by the brand.

How does the CY2027 final rule affect risk adjustment planning?

It changes Stars, not the HCC model. The CY2027 Medicare Advantage and Part D final rule, published April 6, 2026, removes 11 measures from Star Ratings calculations beginning with the 2027 measurement period, keeps the Diabetes Care Eye Exam measure that had been proposed for removal, and adds a depression screening and follow-up measure from the 2027 measurement year that first appears in the 2029 Star Ratings. V28 remains the binding constraint on capture. What the rule changes for a risk-bearing organization is where the same budget goes, because a shorter measure set means each surviving Stars measure carries more of the rating while the risk adjustment work stays exactly as demanding.

See the Risk Adjustment Copilot on your own population.

30 minutes. We bring a working canonical record stitched from your sample feeds, show V28 trumping applied live, and trace a diagnosis from 277CA through MOR. Pelica customers have lifted RAF by roughly +0.4 in two quarters with no new headcount.

Book a Demo → Read the V28 readiness playbook