Page 74

74

than CMS’ proposal to 7.4 percent lower. The range of normalization factors derived from these alternative methods illustrate the inherent uncertainty when predicting future trends. CMS does not believe these alternatives are more reasonable estimates of the average 2026 FFS risk score. CMS is finalizing the proposed normalization factors that were developed using the multiple linear regression methodology that accounts for the different trends in the FFS risk scores between the pre-COVID-19 period and the period during and after by including a COVID-19 indicator for time periods before and after the onset of the COVID-19 pandemic in the regression. This COVID-19 indicator accounts for the distinct difference in the level and year- over-year change in the average FFS risk score between the pre- and post-COVID-19 periods in a way that does not necessitate the need to exclude any years of data. The COVID-19 indicator itself is a categorical and binary variable that identifies in the regression whether an average FFS risk score is based on dates of service before or after the onset of the COVID-19 pandemic. Many commenters based their recommendations on the assumption that the post-COVID-19 trend will return to pre-COVID-19 levels. CMS does not think there is sufficient post-COVID-19 FFS risk score data to support this assumption. The most recent data suggests the FFS risk score trend remains elevated. CMS believes the use of a categorical variable provides valuable transparency in that it is a clear and replicable approach for accounting for the impact of the COVID-19 pandemic in the regression and relies on basic information without the need to make inferences about the impact of certain years in the historical data on future trends. While CMS is finalizing the normalization factors for MA and PACE developed using the multiple linear regression methodology as proposed in the CY 2026 Advance Notice, we will continue to assess trends and the appropriateness of alternate methods for future years. Comment: A commenter stated that their plan experienced a disparate impact during the COVID-19 pandemic compared to others due to their unique population. The commenter recommended that CMS consider the disparate impact of normalization changes and provide stakeholders, especially smaller insurers, additional analyses of the impacts and additional time for consideration of the changes being proposed. Response: We appreciate the request. While CMS acknowledges that individual plan experience and capabilities differ in important ways, the normalization factor is a technical adjustment that accounts for the FFS risk score trend between the denominator year of a model and the payment year. The average FFS risk score changes each year due to an underlying trend that reflects changes in the health status, demographic characteristics, and coding practices in the Medicare FFS population. CMS applies a normalization factor to risk scores in the payment year to account for this trend in the average FFS risk score between the denominator year and the payment year. CMS must predict an average FFS risk score that is a reasonably accurate projection of the FFS risk score in the payment year to maintain an average FFS risk score of 1.0 across the entire FFS population. An appropriate prediction is essential for payment accuracy program wide. Maintaining an appropriate normalization methodology helps to ensure payment stability as routine and necessary updates to the risk adjustment models are made over time. For