The Model That Punishes Volume Over Accuracy
CMS didn’t redesign the HCC risk adjustment model by accident. V28, which reached 100% implementation on January 1, 2026, was engineered to do one thing: reduce the financial reward for coding intensity that doesn’t correspond to genuine patient complexity. Over 2,000 ICD-10 codes lost their HCC mappings. Several HCC categories were eliminated entirely. Coefficients were recalibrated across the board.
The plans hit hardest are the ones that built their risk adjustment strategy around finding the highest number of codes with the greatest RAF impact. Under V24, that strategy worked. Certain diagnosis categories, particularly vascular disease, specified arrhythmias, and diabetes complications, carried outsized coefficients. Aggressive chart review programs targeted those categories specifically because they generated the most revenue per code.
V28 pulled the rug out. Many of those high-value targets now carry significantly less weight. Plans that concentrated their coding effort in those categories are seeing RAF score declines that directly hit revenue, even though their coding volume hasn’t changed. The lesson: volume without clinical accuracy was always a borrowed-time strategy. V28 called in the debt.
Winners and Losers Under the New Model
The coefficient restructuring created clear winners. Plans with genuinely complex patient populations, the ones where chronic kidney disease, heart failure, major depressive disorder, and other high-acuity conditions are well-documented and encounter-linked, are seeing their risk scores hold steady or improve under V28. The model rewards what they were already doing: accurate documentation of real clinical complexity.
The losers are plans whose risk scores were inflated by coding practices rather than patient acuity. Their scores dropped because V28 removed the codes and reduced the coefficients that their programs were designed to chase. The revenue impact is real, measurable, and not going away. There’s no V24 to fall back on. The blended transition ended December 31, 2025.
Finance teams that haven’t recalibrated their revenue models are flying blind. Coding teams working from pre-V28 priority lists are wasting effort on conditions that no longer move the needle. Compliance teams that haven’t assessed how V28 changes their audit risk profile are underestimating their exposure. Every function that touches risk adjustment needs a V28-specific update.
The Compliance Multiplier Effect
V28 doesn’t just change revenue. It concentrates compliance risk. With fewer codes mapping to HCCs, each submitted diagnosis carries proportionally more weight in a plan’s overall risk profile. A single unsupported code has a bigger impact on audit outcomes than it did under V24. The margin for error contracted.
CMS’s population-level pattern analysis amplifies this. When a plan’s coding distribution concentrates in high-value V28 categories, that concentration is visible in the data. It signals targeted coding activity rather than organic clinical documentation. Plans that haven’t adjusted their coding patterns to reflect the new model’s value distribution are painting targets on their own submissions.
The BCBS Alabama OIG audit (A-07-22-01207, March 2026) demonstrated what happens when documentation can’t support submitted codes: 91% error rate, $7.06 million in estimated overpayments, and public findings that become precedent for future enforcement. Under V28, every unsupported code carries more relative weight, making each documentation failure more consequential.
Aligning Strategy to the New Model
Plans that treat V28 as a revenue problem will chase recovery tactics. Plans that treat it as a structural signal will rebuild their programs around accuracy and defensibility. The second group will outperform the first, because V28 isn’t a temporary adjustment. It’s the new permanent baseline.
Thriving under CMS HCC V28 requires updated coding priority lists, two-way retrospective review that adds missed diagnoses and removes unsupported ones, prospective tools that generate encounter-linked codes at the point of care, and explainable AI that builds audit-ready evidence trails for every submitted HCC. The model now rewards clinical accuracy over coding volume. The plans that internalize that shift will find V28 works for them rather than against them.
