Medical Upcoding and Suspecting in Federally Sponsored Healthcare Programs

Exploring the Impact on Risk Adjustment Overpayments in Medicare Advantage and Medicaid

ZeOmega blog series on risk adjustment looks at where we’ve been, what we’re facing today, what’s on the horizon, and how to best prepare. Over the next several blogs, our ZeOmega risk adjustment solution experts will provide a forecast of regulatory changes and guidance on ensuring risk payment accuracy in both business and clinical operations.

The topics we’ll be covering in this series are:

1. Brief overview and history of the Risk Adjustment (RA) program.
2. ‍The results of RA policy and how it’s led to gamification of chronic condition coding.
3. The value of risk-adjusted data across your organization.  
4. How overpayments in risk adjustment threaten the Medicare Trust Fund solvency.
5. Recent federal actions and the likely next steps to control overpayments.
6. Strategy and platform pivots required to move forward successfully in risk adjustment.
7. ZeOmega’s expert tips and how the Jiva Risk Adjustment Navigator meets the evolving RA mandates.

In this second blog, we will cover the results of risk adjustment policy and how it’s led to the gamification of chronic condition upcoding.

Federally sponsored healthcare programs, such as Medicare Advantage and Medicaid, are designed to provide affordable and accessible healthcare to millions of individuals. However, an alarming trend has emerged within these programs - the rising risk of overpayments due to medical “upcoding” and “suspecting”. In this blog, we will delve into the complex world of medical coding, risk adjustment guidelines, and explore how gamification of these practices are leading to increased risk adjustment overpayments in federally sponsored healthcare programs, specifically Medicare Advantage and to a lesser degree, Medicaid.

Understanding Chart Review, Suspecting, and Upcoding

Medical coding serves as the foundation for billing and reimbursement in healthcare. Chart review involves retrospective reviews of medical charts by coders to ensure conditions are coded to the highest specificity. Risk adjustment models include a hierarchy to account for chronic condition complexity and assign higher coefficient values accordingly. CMS will use only the highest level of condition coding in risk score calculations. Coders also make sure that all substantiated diagnosis codes have been submitted for risk adjustment.  However, chart review and its downstream output can lead to violations of Risk Adjustment Data Validation (RAVD) rules, exacerbating the risk adjustment overpayment issue. Common violations from chart review include upcoding and suspecting.

Upcoding refers to the practice of assigning a higher-level code than what is supported by the actual clinical evidence of the condition, leading to increased reimbursement. Suspecting, on the other hand, involves suggesting additional diagnoses based on medical chart review which may not be substantiated by a provider’s documentation. These suspected diagnoses are added to the member’s record but require an encounter and for a provider to attest to the condition by adequately documenting the condition to be submitted for risk adjustment. Unnecessary diagnoses inflate the perceived severity of a patient's health condition, resulting in higher reimbursements. Although these tactics can deviate from CMS guidelines in some cases, they are extremely lucrative for payers.

Risk Adjustment Submissions Requirements

In Medicare Advantage, payers submit encounter data to CMS via the Encounter Data Processing System (EDPS). To meet the requirements for submissions all diagnoses must map to the appropriate HCC (Hierarchical Condition Categories) model for the member. There are multiple HCC models including Part C, Pharmacy, ESRD (End Stage Renal Disease) and PACE (Program of All-Inclusive Care for Elderly Adults); each model provides mapping for diagnosis codes to condition categories along with their Risk Adjustment Factor (RAF) coefficient (representing relative cost per condition) used for calculating the risk score.  Diagnosis codes must also originate from an encounter with an allowable physician type and in an allowable place of service. Lastly, diagnoses must also be substantiated by medical record documentation and are subject to RADV auditing by CMS.

For Medicaid, HCC models vary from state to state but many state programs use the Chronic Illness and Disability Payment System (CDPS) models. Marketplace plans use the Health and Human Services (HHS) risk models published by CMS. In the case of either line of business, submissions are solely based on unmodified encounter data and have similar requirements for code origin and medical documentation. The major difference in MA risk adjustment models and Medicaid or Marketplace is due to population characteristics. For example, MA HCC models do not include pediatric or maternity condition mappings.

CMS Oversight for RA Integrity

CMS oversees the risk adjustment programs and is responsible for providing guidelines to payers and enforcing program integrity audits. These guidelines and auditing procedures are published in the RADV Medical Record Reviewer Guidance document. RADV is a process designed to ensure the accuracy and integrity of risk adjustment submissions in Medicare Advantage. It involves retrospective reviews of medical charts to validate the documented diagnoses and conditions supporting the risk scores. Audits begin with a sample size and violations result in fines extrapolated based on population size and submission period.

While states oversee risk adjustment integrity individually, CMS also oversees programs at a higher level focusing on data quality. Payers submit encounter data to the state first for initial validation, states are then required to submit encounter data to CMS under the Transformed Medicaid Statistical Information System (T-MSIS), which produces a data quality report. In this report, CMS provides error and rejection codes which the state must fix and resubmit. Not addressing these data issues could result in reduced payments.

The Impact of Risk Adjustment Overpayments

Risk adjustment is a crucial component of federally sponsored healthcare programs, ensuring fair compensation for insurers providing care to higher-risk individuals. However, medical upcoding and suspecting have significant consequences, leading to overpayments in the following ways:

1. Inflated Risk Scores: Risk adjustment models rely on accurate documentation of a patient's health conditions to calculate appropriate reimbursements.Upcoding and suspecting artificially inflate a patient's risk score by assigning higher-severity codes or documenting unsupported diagnoses. This leads to insurers receiving higher reimbursements than warranted, resulting in overpayments.
2. Escalating Capitation Rates: Capitation rates, the payments made by the government to Medicare Advantage and Medicaid plans, are determined based on the risk scores of the enrolled population. As risk scores increase due to upcoding and suspecting, capitation rates rise as well. Over time, this trend has resulted in substantial increases in capitation rates, straining the financial sustainability of federally sponsored healthcare programs. The Medicare Payment Advisory Committee (MedPAC) has warned that overpayments threaten the Medicare Trust Fund solvency, which financially supports programs for millions of beneficiaries. In the January 2023 report to Congress , a MedPAC analysis concluded, "Consistent with prior years, nearly all MA contracts had coding intensity greater than [fee-for-service], and the share of MA contracts that are overpaid after accounting for the coding adjustment continues to increase.”
3. Zero Sum Effect - Since state risk adjustment is a “zero-sum game” for the payers in a service area, falsely inflating scores may also disadvantage other payers in the area. Capitation payments are calculated based on a payer’s average risk score compared to other payers in the same service area and market segment. Higher portions of the available funds are paid to payers with higher average risk scores leaving less available funds for payers with lower average risk scores ultimately creating a net zero or zero-sum effect.

Overpayments are less of a threat in Medicaid programs mainly due to the inability to modify or provide supplemental submissions which MA allows. In MA, most of these supplemental submissions are a product of chart reviews and Health Risk Assessment (HRA) data.  

Conclusion

These trends in federally sponsored healthcare programs, most egregiously observed in Medicare Advantage, due to medical upcoding and suspecting are a significant concern.  Moreover, chart review and suspecting can violate RADV rules, further undermining the integrity of risk adjustment calculations.

It is imperative to address these challenges by implementing stricter auditing processes, enhancing education and training on accurate coding practices, and investing in tools based on source of truth data from CMS rather than relying on chart review.  By doing so, we can safeguard the integrity and long-term viability of federally sponsored healthcare programs, ensuring that they continue to provide affordable and quality care to those who need it the most.

Next in our blog series, ZeOmega’s Chief Medical Officer, David Sand, MD, MBA will discuss the value of accurate risk adjustment across organizations.