Can you pass a Risk Adjustment Data Validation audit?
by Yvonne Tso, PharmD, MBA
On January 30, 2023, the Centers for Medicare & Medicaid Services (CMS) published the Final Rule (CMS 4185-F) to identify and recoup overpayments to Medicare Advantage plans (MAOs) by using a statistical extrapolation methodology in risk adjustment data validation (RADV) audits beginning with payment year 2018 but will not use a fee-for-service (FFS) adjuster. The final methodology is grounded in findings reported by the Department of Health and Human Services (HHS), Office of the Inspector General (OIG), the General Accounting Office (GAO), case laws and CMS’ research studies. The estimated amount of recoupment is $4.7 billion over 10 years, approximately 1.3% of the $350 billion CMS paid to MAOs to provide health care coverage for 27 million beneficiaries in PY 2021.
CMS pays MAOs a monthly capitated fee per member enrolled in an MA plan based on risk scores computed from health status and medical conditions of the enrollee. CMS relies on accurate and reliable documentation of health status and medical conditions in the medical records. The obligation to submit properly documented medical records for risk adjusted payment is both a compliance mandate and financial requirement. MAOs must submit data conforming to CMS' requirements equivalent to Medicare FFS data, when appropriate, and to all relevant national standards, including the requirement in the International Classification of Diseases-Clinical Modification (ICD-CM) Guidelines for Coding and Reporting that the medical condition was Monitored, Evaluated, Assessed/Addressed and/or Treated (MEAT) and diagnoses are properly documented in patients’ medical records. When not properly documented. a diagnosis code is not a valid basis for risk adjustment payments. 
Industry standards and guidelines on acceptable or proper documentation in medical records have been published from various sources:
Noridian, a Medicare Administrative Contractor has posted guidelines on its website for FFS providers;
Medicare Learning Network has a Fact Sheet on Complying with Medical Record Documentation Requirements;
The National Committee on Quality Assurance (NCQA) has also provided guidelines; and
CMS has simplified documentation requirements posted on its website
MAOs submit encounter data from multiple sites of care for risk adjusted payments. They should review the data regularly to avoid the following:
Unauthenticated medical records (for example, no provider signature, no supervising signature, illegible signatures without a signature log or attestation to identify the signer, an electronic signature without the electronic record protocol or policy that documents the process for electronic signatures);
Illegibility in writing medical information and each entry should have a legible signature, or the provider's/author's/observer's identity is present and legible to another reader so that a meaningful review may be conducted. Legibility of medical records is not just a billing issue (if the provider is not capitated) or compliance issue; it is a patient care issue. Illegible documentation may result in medication errors and incorrect diagnoses and procedures being assigned to the patient;
Inappropriately altered or addended medical records; medical records cannot be altered. Any errors identified after the original record is complete must be legibly corrected in a manner that allows the reviewer to identify what is being corrected and why. Electronic records should follow the same principle of being able to identify the original entry, the correction, the date and time of the correction, the reason the record is being corrected and the person making the correction.
CMS will use indicators of potential improper payment risk to target MAOs for its upcoming RADV audits from which the government is likely to recoup improper payments. To prepare for or to avoid being a target for RADV audit, MAOs should ensure:
Patient identification is on each page and visit date(s);
Medical information is legible, clear, concise, and reflects patient's condition and includes reason for encounter, relevant history, findings, test results and date of service;
The ability to identify the rendering provider of a service not only for use in standard claims transactions but also for review, fraud detection, and planning policies. The payer (or Medicare) must be able to determine and verify rendering physician/practitioner for each patient service billed. It is very important that individual(s) performing a billed service is/are identified;
Telehealth visits may be documented as a face-to-face visit only when the services are provided using an interactive audio and video telecommunications system that permits real-time interactive communication;
Significant illnesses and medical conditions are indicated on the problem list. CMS looks for a full description of the patient’s condition. A diagnosis may only be coded when it is explicitly named in the medical record. A chronic condition must be restated in the medical record each time it is assessed or treated by the provider and billed on a claim;
Documentation of medications, current and updated (in some cases, medications can support a diagnosis which may have been inadvertently omitted);
Past medical history for patients seen three or more times is easily identified and includes serious accidents, operations and illnesses;
Working diagnoses are consistent with findings;
Only a physician or other qualified and licensed provider legally accountable for establishing a patient’s diagnosis can “diagnose” a patient;
Treatment plans are consistent with diagnoses.
RADV is very time consuming and onerous, especially when medical records are not in electronic format. The financial risk may be raised with extrapolation of audit findings, underreported d
iagnoses from regular monitoring or a RADV audit could result in payment increments. If you would like to discuss how to prepare for RADV, call us at 415 596-5277.
 https://www.cms.gov/newsroom/press-releases/cms-issues-final-rule  UnitedHealthcare Ins. Co. v. Becerra, 16 F.4th 867, 869 (D.C. Cir. 2021), https://oig.hhs.gov/fraud/grant/top-management-performance-challenges, https://www.gao.gov/assets/gao-22-106026.pdf, HHS, FY 2021 HHS Agency Financial Report, pg. 211, https://www.hhs.gov/sites/default/files/fy-2021-hhs-agency-financial-report.pdf., https://www.cms.gov/Research-Statistics-Data-and-Systems/Monitoring-Programs/Medicare-Risk-Adjustment-Data-Validation-Program/Resources.html.  https://www.gao.gov/assets/gao-22-106026.pdf  HPMS Memo April 15, 2022 Reminder of Existing Obligation to Submit Accurate Risk Adjustment Data  (https://www.cms.gov/research-statistics-data-systems/medicare-fee-service-compliance-programs/simplifying-documentation-requirements)  Beginning in 2022, CMS uses 100% encounter data for risk adjustment