CMS finalized sweeping changes to risk adjustment for MSSP and REACH ACOs through the implementation of the V28 CMS-HCC model. Most immediately, the 2023 diagnosis collection year will use a blend of 33% for the V28 model and 67% for the prior V24 model. V28 will be deployed at 67% in 2024 and fully phased in at 100% in 2025 diagnosis collection years. Since these changes have immediate impact, ACOs able to update their V28-specific HCC gap closure strategy mid-year have an advantage. The changes also have implications to PY 2024 early renewal and track decisions, which have already been complicated by the deployment of MSSP program changes in the 2023 Medicare Physician Fee Schedule (MPFS23). Since V28 changes the weighting of specific disease groups, updates to participant selection, specialist engagement, and billing strategies may be needed.
Winners and Losers
ACOs having a higher proportion of patients with conditions weighted lower under V28 compared to V24 are disadvantaged. In some extreme cases, conditions have been removed from V28 entirely. Examples of this are vascular disease and metabolic conditions. Although CMS-HCC models are managed by Medicare Advantage (MA), MSSP takes an additional step to renormalize risk scores relative to national assignable beneficiaries for each enrollment type. These adjustments might have additional effects on V28’s impact based on the ACOs geographic region.
Under V28, coding efforts around cardiology, oncology and nephrology are being rewarded with higher coefficients. For example, the Chronic Kidney Disease coefficients have been increased and differentiated further depending on severity resulting in a positive impact for ACOs. Since ACOs cannot immediately change their attributed patient population, these changes result in winners and losers.
Role of Specialists
The shift in emphasis for these specialties warrant changes to strategies and incentives for engaging specialists, rather than relying only on primary care physicians to close HCC gaps. For some ACOs, over 50% of HCC weight is already being closed by specialists and the percentage is likely to increase with V28.
In cases of specialist practices in the participant list or specialists billing to participant TINs, engagement is often achieved through variants of shared savings distribution. But specialist engagement initiatives and incentives can also work well for owned employed specialists, not just participant specialists attributing patients. Such situations are often helped by EMR and population health system integrations within organizations.
Prioritizing by ROI
Since both V24 and V28 models are in effect during the 2023 and 2024 diagnosis collection years, it is possible to specifically optimize for the exact blended weight that’s in effect. Relying on coding rules and HCC gap suspecting logic in systems designed for the prior V24 model will likely miss many opportunities presented by V28.
In order to make the most of limited resources, ACOs can quantify the impact of coding corrections by simulating the patient with the HCCs gaps closed and recalculating the risk scores. Breaking out the patients by disease cohort and attributed provider, makes it possible to estimate the effort required to intervene with the target patients. Together, these two inputs form the ROI (return on investment) of coding corrections. Plotting the stratified coding accuracy impact by ROI makes it possible to see the curve of diminishing returns. By adding a line that represents the 3% risk ratio capping, it is possible to see in advance when additional efforts yield no improvement. ACOs may need to have two separate plots to cover cases where they renew early into MPFS23 rules in PY 2024 versus continuing in the same agreement period.
Importance of Year-Round Strategy
It is common industry practice for ACOs to review HCC gaps towards the end of the year. This practice often mirrors the HCC sweeps process in Medicare Advantage. But the process of closing HCC gaps for ACOs benefits those who attempt to maintain the same level of coding intensity throughout the year.
It is difficult to “catch up” at the end of the year, even with the most ambitious AWV (annual wellness visit) initiatives. A quantitatively more effective approach is to monitor HCC gap closure rates by PCP, specialist or disease cohort throughout the year. This monitoring establishes a visible feedback loop and can be used to inform the full spectrum of interventions: provider education, EMR alerts at point of care, and provider incentives. Since V28 changes have been finalized in April, ACOs who are able to make mid-year updates to these interventions are at an advantage to those who require more time for these changes.
Impact on Participant Selection
If the ACO includes specialist practices in the participant list or has specialists who bill to participant TINs, there will be an impact on the proportion of attributed patients with specific disease groups. This effect is often more pronounced when advanced practitioners (e.g., NPs) in specialist practices see patients, due to their being classified under “Step 1” level assignment and having higher allowed charges than primary care visits.
As a result, participant selection for involving specialists is the most direct method to improve expected V28 impacts. But in order to have more informed participant selection for PY 2024 that is based on actual contribution to shared savings by participant specialists, the HCC risk scoring forecast calculations need to be updated to use V28. This may in turn result in the proportion of shared savings distributions received by specialists at settlement.
Impact on PY 2024 early renewal and track decisions
The MSSP program does not retroactively recalculate the benchmark year risk scores under model changes. This fact introduces a model version skew into shared savings calculations. For example, an ACO that is advantaged by 9% under V28 may immediately hit the 3% risk cap, despite the V28 model blend being only at 33%. Their benchmark wouldn’t further increase despite improvements in HCC coding accuracy.
One solution to capping from model version skew is to early renew in PY 2024. The agreement period renewal resets the years used for the ACO’s benchmark, enabling it to capture the latest changes in HCC coding improvements. Additionally, since this step deploys the MPFS23 changes to Pathways, the 3% risk ratio cap is calculated in aggregate across all four enrollment types. This has the effect of making it less likely that an ACO’s risk score will be capped.
That said, ACOs not ready for downside risk should recognize that the early renewal option may accelerate their glidepath into a higher risk track. The exact impact of this depends on the number of years remaining in the upside only tracks of the current agreement period. The more years remain, the more upside-only years are given up with early renewal.
V28 changes require an update to HCC risk score forecasting, which is possible due to the diagnosis collection window lagging the risk score application by one year. These forecasts in turn change to the simulated probability of shared savings for future years. It is possible that a change to a less favorable forecast might cause some ACOs to remain longer in lower risk tracks.
The introduction of the new V28 CMS-HCC Risk Adjustment Model could be beneficial for some ACOs and problematic for others. But although some ACOs may start out being relatively disadvantaged, the outcome is not a foregone conclusion. ACOs have the ability to take advantage of the changes even in 2023 by leveraging specialist engagement, prioritizing coding corrections by ROI, and implementing year-round HCC gap closure initiatives. They also have the ability to improve their long term performance starting PY 2024 through participant selection, evaluation of early renewal, and track decisions.