Factors are aligning that may impact requirements for reporting conditions in the Medicare Advantage program using Hierarchical Condition Category (HCC) coding. CMS has recently taken steps through the rule-making process to increase penalties for reporting conditions that are not supported by the clinical record. The agency has also updated the CMS HCC model used to determine risk adjustment factor coefficient (RAF) scores to reduce what it refers to as “gaming.” A logical next step for CMS would be to raise the documentation bar to require that diagnoses are “clinically validated” when reported.
Clinical validation is a higher level of coding that looks at the entire clinical picture to confirm whether a diagnosis is adequately supported or if a procedure is warranted. Recent Medicare Advantage audit reports published by the Office of the Inspector General (OIG) of the U.S. Department of Health and Human Services (HHS) suggest that the OIG may be moving toward clinical validation when determining the validity of a reported condition. The OIG invalidated reported diagnoses based on what could be interpreted as clinical criteria (as opposed to coding criteria). Examples could include the lack of an antidepressant medication being prescribed for a patient with a reported, major depressive disorder, and the use of a medication that treats neurogenic claudication as opposed to vascular claudication. The reports do not give a complete picture of how invalidation decisions are made, but they do appear to be using clinical criteria to invalidate certain diagnoses.
This could have significant impact on reporting conditions in the Medicare Advantage model. There is a general lack of specific CMS guidance for the amount and type of supporting documentation needed to validate a diagnosis. As a result, many Medicare Advantage Organizations (MAOs) have developed their own unique documentation requirement policies for supporting documentation. These policies may include specific instruction regarding the types of clinical indicators needed to support reported diagnoses, and some may allow for reporting of a condition if it is supported by a single clinical indicator.
However, there is no official guidance whether a single clinical indicator or multiple clinical indicators are needed when reporting conditions. CMS has also not explicitly defined the term “supporting documentation,” at least not in a way that guidance can be extrapolated to many common real-world documentation scenarios.
Current guidance available in the CMS 2008 Risk Adjustment Participant Guide and the ICD-10 CM Official Guidelines for Coding and Reporting can be variously interpreted as meaning that evidence of evaluation and treatment or management meets the requirement for supporting documentation when conditions are reported. Given available guidance is nonspecific, should MAOs start using clinical validation, i.e., identify multiple clinical indicators that ensure that the diagnosis is clinically valid prior to reporting conditions, or are current policies adequate?
Confusing matters further, the Official ICD-10-CM Guidelines for Reporting and Documentation, Guideline Section I (A. 19) states that the “assignment of a diagnosis code is based on the provider’s diagnostic statement that the condition exists. The provider’s statement that the patient has a particular condition is sufficient…”
The term “diagnostic statement” is not defined but could reasonably be interpreted as the assessment of an encounter note or the discharge diagnoses section of a hospital discharge summary but does not specifically exclude diagnostic statements in other sections of the record. This guideline could also be reasonably interpreted to mean that a diagnosis reported in a diagnostic statement without clinical indicators would be sufficient from a coding standpoint, however this would clearly not meet the requirements for clinical validation.
In the event CMS adopts clinical validation as a requirement for reporting diagnoses for Medicare Advantage, it would be advisable to establish clearly defined requirements and a future effective date. This will allow MAOs to update their documentation policies to include clinical validation of reported diagnoses, which in general requires that diagnoses are supported by a wider range and number of clinical indicators than is currently required by many MAOs.
Clinical validation is significantly more labor intensive than traditional coding and requires deeper clinical knowledge. This may be offset by the growing use of artificial intelligence technologies that can accurately identify a multitude of clinical indicators for each diagnosis, putting the coding professional in the role of agreeing or disagreeing with the computer’s findings. Clinical validation is also being increasingly used to deny claims and may evolve into a touchstone for other reimbursement models. MAOs should consider how technology that leverages clinically tuned natural language processing and machine learning can increase their efficiency without introducing error.
References
- Official ICD-10-CM Guidelines for Coding and Documentation.
- Poe Bernard, S and Singh, K. “Capture the Complete Clinical Picture With Precision” AAPC Blog 2022.
- Poe Bernard, S. Risk Adjustment Documentation & Coding, 2nd Edition, 2020.
About the author
Michael Stearns, M.D., CPC, CRC, CFPC is a physician informaticist, health information technology (HIT), and health care compliance professional. He has over 20 years of experience in the areas of electronic health records (EHRs), telehealth, virtual services, quality reporting, risk adjustment, health information exchange, clinical terminology development, standards, and billing and coding compliance.
Dr. Stearns has provided leadership to informatics and terminology projects at the National Library of Medicine, the National Cancer Institute, and the College of American Pathologists. He served as the International Director of SNOMED International during the development of SNOMED CT and he was the founding board president of the Texas e-Health Alliance. Dr. Stearns is the co-founder and a lecturer at the University of Texas at Austin Health Information Technology Certificate Program.
Dr. Stearns co-chaired the American Health Information Management Associations’ (AHIMA) Technology and Innovations Practice Council in 2020. He also served as a subject matter expert for the Health Information Management Department of Grand Canyon University. Dr. Stearns has won several awards for teaching, patient privacy and EHR-related patient safety. He is currently the specialized consulting director–medical informatics at Wolters Kluwer, Health Language.