Diagnosis coding practices in the Medicare Advantage (“MA”) risk adjustment space continue to be a top enforcement priority for the Department of Justice (“DOJ”). Last month, DOJ announced the latest such resolution: a $14.1 million settlement with Complete Health Partners Holdings, LLC (“Complete Health”) to resolve allegations that it caused the submission of false diagnosis codes in violation of the False Claims Act (“FCA”). This settlement follows a series of recent MA risk-adjustment enforcement actions across the industry, from health plans to management services organizations to in-home assessment vendors, and offers a useful window into DOJ’s current thinking on programs designed to generate diagnosis coding suggestions, particularly those programs that enter suggestions directly into the electronic medical record (“EMR”) or focus on a particular set of billing codes.
The Settlement. The underlying case, United States ex rel. Bowers v. Complete Health Partners, Inc., et al., No. 3:22-cv-463 (M.D. Fla.), was brought under the qui tam provisions of the FCA by a former employee of the two Medicare Advantage Organization (“MAO”) co-defendants. The United States intervened in the case only to settle certain claims against Complete Health. The claims against the two MAO co-defendants and Complete Health’s private equity owner and operator Pharos Capital Group, LLC will be dismissed without prejudice as to the United States.
The Government contends in the settlement agreement that Complete Health “pressured” providers to add diagnosis codes within Hierarchical Condition Code (HCC) 55 (drug and alcohol dependence) and HCC 59 (major depressive, bipolar, and paranoid disorders) that were not clinically valid, improperly supported by the beneficiary’s medical records, or not considered in the beneficiary’s care, management, or treatment. Specifically, coders inserted specific diagnosis codes within HCC 55 and HCC 59 directly into the EMR as “suggestions” before a beneficiary’s visit, prompting providers to add those codes. The allegedly fraudulent addition of these codes inflated patients’ risk scores, leading to increased future capitation payments under Medicare’s risk adjustment payment model.
The Broader Enforcement Trend. Complete Health is far from the only entity to face government scrutiny for risk adjustment coding practices. Over the past several years, DOJ and relators have pursued a series of FCA cases against MA plans, provider groups, and vendors involved in the risk adjustment process, many of which have involved high-risk diagnosis codes the same as or similar to those at issue in the Complete Health settlement.
- At the beginning of this year, DOJ entered into a $556 million settlement agreement with a health system related to, among other things, an alleged practice of suggesting additional diagnosis codes after a patient encounter.
- Earlier this summer, DOJ announced a $56.5 million settlement agreement with an in-home health assessment vendor to resolve allegations that the vendor caused MAOs to submit invalid diagnoses of various chronic conditions including drug dependence, major depressive disorder, and morbid obesity. The Government alleged that these diagnoses were documented without clinical confirmation, were based only on patient attestations and history, and frequently were not diagnosed by any other healthcare provider during the same time period.
- Also this year, a national health insurer agreed to pay $117.7 million to resolve allegations that it operated a chart review program that added diagnosis codes it determined to be supported, but did not delete unsupported codes. A portion of the settlement was specifically directed toward allegations concerning the addition of diagnosis codes for morbid obesity for patients without qualifying BMIs.
OIG Guidance. As Sheppard previously covered, the Department of Health and Human Services Office of Inspector General (“OIG”) in February published the Medicare Advantage Industry Segment-Specific Compliance Program Guidance, its first Medicare Advantage-specific guidance since 1999. That guidance specifically refers to the use of prompts to add risk-adjusting diagnoses within EMRs as “potentially fraudulent and abusive conduct”.
Implications for Programs that Generate Coding Suggestions. Although these settlements indicate government interest in platforms that generate diagnosis coding suggestions, they should not be read as an indictment of all such practices. Rather, this trend reemphasizes the importance of accuracy and thorough documentation as the cornerstones of any risk adjustment program.
Critically, the Government linked its concerns about Complete Health’s coding suggestion process to its alleged effect on the adoption of inaccurate diagnosis codes. This point is important, since it is not enough in an FCA case for DOJ to make allegations about coding-related processes – rather, it needs to show that the processes resulted in actual false claims. Therefore, the quality of a system’s coding suggestions, and of review and adoption of suggestions, will necessarily drive risk. Ensuring that suggestions reflect diagnoses that are supported by the medical record – not the other way around – and that they are substantively reviewed by a provider before being accepted is one safeguard against FCA liability.
Diagnosis Codes of Interest. The diagnosis codes highlighted by these settlement agreements are also notable as codes that may be especially subject to documentation challenges. For example, some of the enforcement actions focused on substance dependence and mental health diagnosis codes in HCC 55 and HCC 59, obesity, arthritis, and other conditions such as congestive heart failure, arrhythmias, and thrombophilia. These codes are frequent targets of DOJ scrutiny and can be difficult to document and apply in senior populations. Providers and plans may wish to consider additional oversight with respect to these codes to ensure that they are fully supported by the medical record and that clinicians are appropriately following up on identified diagnoses.
Conclusion. The wave of recent MA risk adjustment enforcement and OIG’s new MA compliance guidance provide an additional window into the Government’s approach to coding suggestion and provider query tools, difficult-to-substantiate diagnoses, and substance use disorder and behavioral health diagnosis codes. For providers, management services organizations, and MAOs that may use diagnosis coding suggestion platforms or programs, these developments highlight the central importance of a program’s quality of diagnosis suggestions and the robustness of provider review. In light of these trends, organizations operating similar programs in the MA space may want to evaluate whether suggested diagnosis codes are clinically supported, whether provider review is documented and substantive, and whether coding-related incentive structures are designed to reward accuracy rather than volume.