Today we’re joined by James Maitland, an expert in robotics and IoT applications in medicine, who brings a unique, technology-driven perspective to the complex world of healthcare policy. With Medicare Advantage now covering over half of all Medicare enrollees and estimates of federal overpayments reaching a staggering $76 billion this year alone, the system is under intense scrutiny. We’ll explore the mechanics behind these overpayments, discussing practices like “upcoding” and the impact of recent regulatory fixes. Our conversation will also delve into the fundamental payment incentives that drive insurer behavior, the controversies surrounding how costs are calculated, and the potential fallout from increased government investigations into billing practices.
Estimates suggest Medicare Advantage overpayments will total $76 billion this year. Can you explain the mechanics behind “upcoding” and “favorable selection” and how each practice contributes to MA plans costing significantly more than traditional Medicare for the same enrollee? Please provide some real-world examples.
It’s a fascinating and deeply flawed system from a data-incentive standpoint. Think of it this way: the government pays private MA plans a flat, risk-adjusted fee per member. “Upcoding” is the direct result of that incentive. Plans have a powerful financial motivation to make their members appear as sick as possible on paper, because a higher risk score means a higher payment. For example, a plan’s software might scan a doctor’s notes, find a mention of a past, resolved condition, and add that code to the member’s current file, boosting their risk score and the plan’s revenue, even if that condition requires no active treatment. Then you have “favorable selection,” which is the other side of the coin. MA plans, with their gym memberships and other wellness perks, tend to attract healthier, less costly seniors. So, plans receive payments based on an average risk profile but are covering a population that uses less care. The MedPAC report quantifies this perfectly: without these two factors, MA spending would actually be 1% less than traditional Medicare. But favorable selection adds 11 percentage points to their cost, and upcoding adds another 4, bringing MA spending to 114% of what it would be in the traditional program.
A new risk adjustment model, V28, has been implemented to curb overpayments, yet some describe it as a “blunt tool.” In what specific ways does an across-the-board adjustment negatively affect smaller, regional plans that may not be upcoding, and how does it impact their ability to compete?
The V28 model is a classic example of a broad-stroke solution that creates unintended consequences. It’s designed to reduce payments system-wide to counteract the aggregate effect of coding intensity. On one hand, it has been effective; we saw overpayment estimates drop from a projected $84 billion to $76 billion. The problem is that it doesn’t distinguish between a plan that is aggressively gaming the system and one that is coding accurately. It’s a penalty applied to everyone. For a smaller, regional plan that has been playing by the rules, this across-the-board cut feels incredibly unfair. It directly reduces their revenue, which in turn limits their ability to fund the supplemental benefits—like dental, vision, or transportation—that are crucial for attracting and retaining members. This puts them at a significant competitive disadvantage against the massive national insurers who have more financial cushion and are, according to the data, the ones who benefited most from intense coding in the first place.
It’s been argued that the MA payment system creates “irresistible incentives” for plans to focus on maximizing reimbursement through coding. What fundamental changes to this payment chassis could realign incentives toward better patient management and reducing the actual cost of care, as was the original promise of MA?
This gets to the heart of the matter. As one commissioner put it, the incentives to “play the game” are simply irresistible. The profitability of coding has become so immense that it has overshadowed the actual mission of managing patient care effectively. The original promise of MA was that these private plans, with their technological and managerial efficiencies, could deliver better care for less money. Instead, they’ve become experts at administrative optimization for profit. To fix this, you have to change the “underlying payment chassis,” as Commissioner Lynn Barr described it. We need to move away from a system that rewards diagnoses on paper and toward one that rewards tangible outcomes and actual cost reduction. This could mean incorporating more quality metrics, patient satisfaction scores, and, most importantly, measures of whether a plan is actually reducing hospitalizations or lowering the total cost of care for a given condition. The focus must shift from “how sick is this patient?” to “how well are we keeping this patient healthy?”
The methodology used to calculate overpayments often focuses on beneficiaries who switch from traditional Medicare to MA. What are the main criticisms of this approach, and what alternative data or analytical methods could provide a more universally accepted comparison of costs between the two programs?
The methodology is a major point of contention, and frankly, the critics have a point about its limitations. The MedPAC approach analyzes the health spending of people before they switch from traditional Medicare to MA and compares it to their peers who stay. The data consistently shows these “switchers” are healthier and use fewer services. The criticism, led by industry groups like the Better Medicare Alliance, is that this is a “narrow and biased sample.” They argue it doesn’t account for people who are new to Medicare entirely or those who have been in MA for a long time. They believe this focus exaggerates the “favorable selection” effect and fails to capture the full picture. To get a more universally accepted comparison, we’d need much better data transparency from the MA plans themselves. This includes more granular data on encounter rates, the cost of supplemental benefits, and long-term health outcomes of their enrollees. An ideal model might use a synthetic control group, matching MA enrollees with statistically identical counterparts in traditional Medicare across a wide range of variables, not just focusing on the single act of switching programs.
With the Department of Justice actively investigating and settling with major insurers over billing practices, how is this heightened regulatory scrutiny likely to alter plan behavior? Beyond fines, what specific operational or compliance changes do you anticipate insurers will implement in response?
The increased scrutiny from the Department of Justice is definitely turning up the heat. A fine, even a massive one like the record $556 million settlement with Kaiser Permanente affiliates, can sometimes be treated by a large corporation as just a cost of doing business. However, the threat of ongoing investigations and the associated reputational damage is a much stronger motivator for change. I anticipate we’ll see insurers investing heavily in internal compliance and auditing systems. They’ll likely implement more robust chart review protocols and AI-powered tools designed not just to maximize coding, but to validate it against clinical evidence to avoid crossing the line into fraud. Operationally, this means hiring more compliance officers, conducting more rigorous training for their coding staff, and creating stricter internal guardrails on what can and cannot be submitted for risk adjustment. It forces a cultural shift, however slow, from “what can we get away with?” to “what can we defend under a federal microscope?”
What is your forecast for the Medicare Advantage payment model over the next five years?
My forecast is that we are heading for a significant recalibration. The current model, with its massive overpayments and misaligned incentives, is becoming politically and fiscally unsustainable. The $76 billion figure is simply too large to ignore. Over the next five years, I expect to see a multi-pronged approach from regulators and Congress. First, risk adjustment models like V28 will continue to be refined, becoming more sophisticated and targeted to penalize outlier coding behavior rather than applying blunt, across-the-board cuts. Second, there will be a much stronger legislative push for data transparency, forcing plans to open their books on spending and outcomes. Finally, and most importantly, I believe we will see the beginning of a fundamental shift in the payment chassis itself—a move toward value-based payments that reward plans for demonstrably improving health and lowering total costs, rather than just for being good at paperwork. The program won’t be scrapped, but the pressure to align its payments with its original promise of efficiency and better care will be immense.
