Build Defensible RAF Scores: 2026 Strategy Guide

What Are Defensible RAF Scores?

A risk adjustment factor (RAF) score predicts the expected cost of a Medicare Advantage enrollee. A score of 1.0 represents the average Medicare beneficiary; above 1.0 indicates higher expected costs and higher reimbursement [2].

Most organizations treat RAF scores as an output of coding, separate from the risk scores that drive payments. That’s the wrong frame. Defensible RAF scores are the output of clinical documentation, evidence validation, and process integrity.

A defensible score means every diagnosis submitted answers three questions CMS demands proof of. Where in the medical records is this condition documented? Does it meet MEAT criteria (Monitoring, Evaluation, Assessment, Treatment)? Can you trace the chain from patient encounter to submitted claim?

Coding accuracy tells you a code is correct. Defensibility proves why. That distinction separates surviving a RADV audit from facing repayment demands.

Why Defensible RAF Scores Matter More in 2026

CMS Is Auditing Every Eligible Plan

In May 2025, CMS announced an accelerated audit strategy, expanding from roughly 60 contracts per cycle to all eligible Medicare Advantage plans annually [3]. The January 2026 memo confirmed Payment Year 2020 audits are underway, with new audits initiating quarterly and sample sizes up to  200 enrollee records [4]. For MA organizations, every contract now faces scrutiny.

The Court Ruling Didn’t Remove Risk

A September 2025 federal court vacated the RADV Final Rule that would have allowed CMS to extrapolate error rates across entire contract populations [5]. But CMS still collects overpayments on sampled records, the ruling may be appealed, and CMS has stated that RADV audits remain a top priority [4].

Coding Intensity Is Under Scrutiny

MedPAC’s March 2025 report estimates that favorable selection and coding intensity together will generate approximately $84 billion in excess Medicare Advantage payments, with coding intensity alone accounting for roughly $40 billion [6]. The FY2025 Part C improper payment rate was 6.09% ($23.67 billion), with most attributed to documentation that failed to substantiate diagnosis data [7]. OIG targeted audits continue finding that medical records did not support submitted codes in the majority of sampled cases, and errors like these remain a primary concern for the Centers for Medicare and Medicaid Services [8].

V28 Raises the Documentation Bar

Payment Year 2026 marks the full implementation of the CMS-HCC V28 model [9]. V28 eliminates over 2,000 diagnosis codes previously mapped to hierarchical condition categories, expands categories from 86 to 115, and recalibrates weights for chronic conditions, including diabetes and its chronic complications [10]. The OIG has announced a study comparing V24 and V28 coding patterns to track organizational adaptation [10].

Five Principles of Defensible Coding

1. Link Every Diagnosis to a Patient Encounter

CMS requires that codes reflect conditions documented during a face-to-face encounter within the payment year [11]. Diagnosis capture from chart reviews or health risk assessments without a corresponding provider visit is the highest-risk category in audits.

2. Meet MEAT Documentation Standards

Each diagnosis needs documentation showing a provider is actively Monitoring, Evaluating, Assessing, or Treating the condition. Listing a chronic disease in the problem list without a clinical context doesn’t meet the standard.

3. Revalidate Chronic Conditions Annually

CMS resets risk scores every year. Chronic conditions like heart failure, COPD, major depressive disorder, and diabetes with complications must be re-documented during qualifying encounters. Without current-year documentation, revenue is indefensible.

4. Justify Severity

Under V28, severity coding matters more than ever. Diabetes without complications no longer maps to an HCC, while diabetes with chronic complications carries meaningful value. Clinicians must document specific severity and supporting evidence in each clinical note.

5. Maintain Data Integrity

Defensibility requires organized, retrievable records with traceable logic. Which coder reviewed the chart? What clinical evidence supported the code? When was the evaluation completed? Health plans that can’t answer these questions carry unquantified liability.

Two Common Defensive Coding Techniques

Two-way retrospective review. Most retrospective programs only find missed codes. This add-only approach raises flags for regulators. Defensible programs also identify and remove unsupported diagnoses that lack encounter linkage or current-year documentation. A program that never deletes signals revenue intent, not clinically accurate coding [1].

Prospective documentation and coding support. The safest diagnosis is one confirmed during a live encounter. Prospective programs provide clinicians with pre-visit summaries and clinical decision support at the point of care, producing encounter-linked documentation by default.

What Is a Normal RAF Score?

A score of 1.0 represents the baseline Medicare beneficiary [2]. Below 1.0 suggests a healthier population; a value above 1.0 indicates higher expected costs. There’s no universal “normal” because the correct score depends on each patient’s actual disease burden.

For context, the 2025 base rate is approximately $10,402 per member per year [12]. Every 0.1 increase translates to roughly $1,040 in additional annual reimbursement. For a plan with 100,000 members, even a 0.05 improvement across the population represents over $5 million annually.

Is RAF the Same as Risk Score?

Related but technically different. The risk score is the raw CMS-HCC model output from demographics and mapped diagnosis codes. The risk adjustment factor is the final adjusted number after CMS applies the normalization factor and the MA coding intensity adjustment [13]. In practice, both terms refer to the score that drives per-member reimbursement for Medicare Advantage plans.

What Is RAF Scoring?

RAF scoring is how the Centers for Medicare and Medicaid Services assigns a risk adjustment factor to each enrollee. CMS maps ICD-10 codes to hierarchical condition category (HCC) groupings, assigns coefficient weights, and combines those with demographic factors to produce the capitated payment amount for health plans and provider groups [11]. The V28 model, fully implemented for 2026, uses updated cost data and tighter clinical groupings to improve predictive accuracy [9].

What Is the RAF Score for Heart Failure?

Under V28, heart failure maps to HCC 224 with meaningful RAF weight. The exact coefficient depends on the clinical setting and disease interaction terms when heart failure co-occurs with other chronic conditions like diabetes or kidney disease [14]. For example, the combination of heart failure and diabetes generates an additional interaction coefficient. Documentation must capture severity, clinical status, and evidence of active care management to ensure compliance and defensibility.

How Coding Intensity Becomes Compliance Risk

Coding intensity is the pattern where MA plans document more codes for the same population than traditional Medicare would [6]. The V28 model was designed to reduce the impact of discretionary coding on risk scores [10]. CMS applies a uniform adjustment across all plans, meaning organizations with clinically accurate coding are penalized alongside those gaming the system [6].

The shift from coding intensity to coding integrity isn’t optional. The DOJ’s Kaiser settlement, OIG’s ongoing audits, and CMS’s accelerated RADV schedule all point in one direction: more codes without proportional evidence means more fraud exposure on the balance sheet.

How Neuro-Symbolic AI Supports Defensible RAF Scores

Traditional NLP systems find codes by scanning clinical notes for diagnosis language. What they don’t do is prove why a code is valid.

Neuro-symbolic AI combines neural network pattern recognition with a knowledge-driven symbolic reasoning layer built on medical rules, ICD-10-CM guidelines, and MEAT criteria logic. Every recommendation includes a transparent evidence trail: the specific text in the note, the MEAT elements present, and the reasoning chain linking documentation to the suggested code.

For health plans and provider groups, this transforms the economics of RAF scoring. Chart reviews that once took 40+ minutes can be processed in under 8 minutes while maintaining over 98% coding accuracy.* Neuro-symbolic AI also flags unsupported diagnoses for removal and runs audit simulations to identify potential issues before submission.

*RAAPID internal benchmark.

Building Your Defensible RAF Strategy

Centralize operations. Replace fragmented spreadsheets and siloed systems with real-time visibility into coding status, documentation quality, and audit readiness.

Implement a two-way retrospective review. Capture missed codes and remove unsupported ones. Track the add-to-delete ratio as a key metric.

Invest in provider education. Build programs that help clinicians document conditions completely during encounters, including severity, chronic complications, and social determinants of health.

Validate before submission. Run mock audits using CMS logic. Flag weak charts, remediate gaps, and ensure every code has defensible support.

Align stakeholders. Connect compliance, finance, operations, and IT around shared dashboards and metrics.

What Comes Next

Risk adjustment has shifted from a revenue function to a clinical care, compliance, and enterprise AI discipline. Health plans that build defensible RAF scores on encounter-linked documentation and transparent validation will capture every dollar they’re legitimately owed. The revenue isn’t missing. It’s waiting for organizations with the confidence to claim it.