Overview
MEAT is a documentation framework used widely in risk-adjustment coding compliance to demonstrate that a chronic condition was actively addressed during a face-to-face encounter. The acronym stands for Monitor, Evaluate, Assess, and Treat — a documented presence of any one element is generally sufficient to support ICD-10-CM coding of the condition for that encounter.
Monitor means the condition's status is being tracked. Documentation examples: "diabetes stable per today's review," "HbA1c 7.2, continuing current regimen," "BP today 128/78, at goal." Evaluate means the condition is being assessed through examination, testing, or review of results. Examples: "reviewed CBC and BMP," "lung exam reveals scattered wheezes," "foot exam unremarkable." Assess means clinical decision-making about the condition. Examples: "COPD moderate per GOLD criteria," "CKD stage 3b based on today's eGFR." Treat means active therapy including medication management, referral, or procedure. Examples: "continue metformin 1000mg BID," "increase lisinopril to 20mg," "referred to pulmonology for COPD exacerbation workup."
MEAT exists because CMS Risk Adjustment Data Validation (RADV) audits and private-payer audits consistently recover overpayments for diagnoses that appear in claims but lack current-year documentation support. The CMS Medicare Managed Care Manual Ch. 7 and supporting sub-regulatory guidance require that risk-adjusting diagnoses be supported by documentation in the medical record showing the condition was addressed during a face-to-face encounter in the data collection year. MEAT operationalizes this requirement into a teachable, auditable framework.
For revenue cycle and coding leaders, MEAT is the core content of risk-adjustment CDI training. Physicians often inherit problem lists from specialists or prior PCPs, and routinely document "continue current management" without MEAT language. Structured training teaches providers to add one MEAT element per chronic condition per visit — a 30-second documentation augmentation that prevents RAF loss. EHR templates and risk-capture prompts that surface prior-year HCCs with single-click MEAT language accelerate adoption.
MEAT has limits. It does not substitute for clinical accuracy — a condition must actually exist and meet diagnostic criteria. Adding "monitor" language to a diagnosis that is not supported by labs, exam, or patient-reported symptoms is not compliant even if MEAT is documented. RADV auditors review the full chart context, not just the MEAT language. And some specialties operate in encounters too brief for fulsome MEAT documentation; annual wellness visits, comprehensive annual exams, and chronic-care-management touchpoints are typically the best venues for structured risk capture.
Many organizations extend MEAT to TAMPER (Treatment, Assessment, Monitor, Plan, Evaluate, Referral) or other mnemonics. The specifics matter less than the principle: current-year face-to-face documentation showing the condition was addressed clinically, not merely listed on the problem list. Programs that systematize MEAT in provider training, EHR templates, and pre-visit chart preparation routinely capture 15–25% higher accurate RAF than those relying on passive problem-list coding.
Industry benchmark
CMS Medicare Managed Care Manual Ch. 7 sections on documentation requirements. CMS RADV audit methodology (published annually). Industry reference: AHIMA and AAPC risk-adjustment coding guides.
Worked example
A 68-year-old Medicare Advantage member with diabetes, hypertension, and COPD has an annual wellness visit. Provider documentation under MEAT framework: Diabetes — 'HbA1c 7.1, stable, continue metformin' (Monitor + Treat). Hypertension — 'BP today 134/84, slight elevation from goal, increasing lisinopril to 20mg' (Evaluate + Treat). COPD — 'mild wheezing, pulmonary function stable, continue Trelegy, recent COPD exacerbation in February managed with prednisone' (Monitor + Treat). All three HCCs supported; estimated RAF impact captured: +0.88 for the year.
Frequently asked questions — MEAT Criteria
Does every diagnosis need MEAT documentation?
For chronic conditions being coded for risk adjustment, yes — each diagnosis needs at least one MEAT element documented during a face-to-face encounter in the data collection year. Acute conditions that resolve within the encounter are documented through the standard assessment and plan; MEAT is specifically for chronic-condition carry-forward.
Is 'patient takes metformin' sufficient MEAT documentation?
Marginal. Medication listings alone without clinical context ('continuing metformin for diabetes' with any assessment of control) weaken the MEAT evidence. Best practice adds one additional element: 'continuing metformin, HbA1c stable at 7.2' or 'on metformin, reviewed for side effects, no hypoglycemia.'
Can telehealth visits support MEAT documentation?
Yes, within CMS-defined telehealth parameters. CMS permits telehealth-based diagnosis capture for risk adjustment through current flexibilities extending from the Public Health Emergency. However, lab-dependent monitoring may be limited in virtual encounters; comprehensive in-person annual wellness visits remain the highest-yield MEAT venue.
What's the auditor's standard for MEAT sufficiency?
RADV auditors review the complete encounter for reasonable clinical documentation supporting each coded condition. A single explicit MEAT element per condition is typically sufficient if the clinical context supports the diagnosis. Auditors flag problem-list-only coding, copy-paste documentation without new clinical observation, and diagnoses inconsistent with the encounter's presenting problem.
Disclaimer
This glossary entry is operational reference for revenue-cycle and medical-billing professionals. It is not legal, clinical, or contractual advice. Industry benchmarks cite named public sources where available; always verify against the current guidance from the authority body before relying on a number in a contract, policy, or compliance filing.