Summary/Response to Public Comment

CMS Response to Public Comments Received for CMS-10203

CMS received two comments related to CMS-10203 (OMB control number: 0938–0701) for the Medicare Health Outcomes Survey (HOS).

• Comment: One commenter supports CMS’s proposed removal of six questions to make the survey shorter but asked that CMS evaluate and share the impact on case-mix adjustment. Specifically, the commenter is concerned the three common chronic medical conditions, which they note are not specifically related to Star Ratings, could have an impact on the death case-mix model portion of the case-mix adjustment for Improving or Maintaining Physical Health (PCS). This commenter also suggested that CMS explore removing the death case mix model from case-mix adjustment.

Response: In 2020, CMS evaluated the contribution of several items to the case-mix adjustment of the HOS longitudinal measures (Improving or Maintaining Physical Health and Improving or Maintaining Mental Health), including the three chronic condition indicators (arthritis of the hip or knee, arthritis of the hand or wrist, and sciatica) currently proposed for removal. The standard deviation of changes associated with case-mix adjustment summarizes the effect of case-mix adjustment by showing the typical magnitude of change. The standard deviations in the Improving or Maintaining Physical Health and Improving or Maintaining Mental Health scores associated with removing these case-mix adjusters are ~0.2 points on a 0-100 scale, accounting for approximately 0.5-1% of case-mix adjustment variance. Removing these items from the HOS instrument would therefore have little impact on case-mix adjusted scores.

Removing these three common chronic medical condition questions would reduce respondent burden while resulting in negligible and non-substantive changes in scores. The proposed changes to the survey questionnaire would be implemented in 2022 and would not impact case-mix adjustment for the longitudinal HOS measures before the 2026 Star Ratings.

We appreciate the commenter’s suggestion that CMS explore removing the death case-mix model from HOS case-mix adjustment methodology. CMS will consider this suggestion as we continue to explore new ways to enhance the HOS and develop less complicated measures.

• Comment: One commenter expressed concern about Physical Functioning Activities of Daily Living (PFADL), a new longitudinal change measure derived from the HOS, and the added burden for beneficiaries and health plans that it presents. The commenter also believes the wording of the component questions sets up a pre-emptive bias against beneficiaries who are disabled and those who have permanent functional limitations, and that it allows little opportunity for health plans that enroll individuals with disabilities (SNPs) to show improvement on the measure.

Response: First, we would like to clarify that the PFADL change score measure currently is a Star Ratings Display measure and that it has not been proposed as a Star Ratings measure through the regulatory process at this time. Also, all eight component questions of the new measure (two physical functioning questions and six ADL questions) are already collected by the HOS so the measure does not entail any additional data collection burden for plans or increased response burden for beneficiaries.

CMS’s intent in the development of this measure is to use the existing HOS and HOS-M ADL items, including those that are currently used for frailty assessments for Program of All-Inclusive Care for the Elderly (PACE) and Fully-Integrated Dual Eligible Special Needs Plans (FIDE-SNP). The commenter believes the wording of the ADL questions (“…do you have any difficulty doing the following activities without special equipment or help from another person”) allows little opportunity for plans that enroll individuals with disabilities to show improvement on the measure. It is not clear whether basic, nonprescription devices, such as railings, grab bars, and shower chairs, are likely to be interpreted as “special equipment” when beneficiaries respond to these questions. However, we will explore this issue further as we continue to develop the measure. Specifically, CMS will consider cognitive testing of the ADL items among a small number of enrollees to better understand how beneficiaries interpret that phrase.

Declines in MAO beneficiary physical function predict disability, institutionalization, and mortality, but with early detection and appropriate intervention, such declines may be preventable and even reversible. Therefore, CMS remains committed to measuring MAOs’ contributions to the physical functioning of MAO beneficiaries using new measures such as the PFADL change score. CMS also recognizes that there is variation in baseline function across MAO contracts at enrollment and therefore case-mix adjusts the PFADL change score measure for baseline function scores. By doing so, there is no advantage to contracts with higher or lower functioning beneficiaries at enrollment or baseline.

CMS appreciates the commenter’s feedback on the new PFADL measure. As part of the continuing measure development, CMS is planning additional analyses to explore other possible beneficiary-level case-mix adjustors related to SNP status, which may include dual eligibility and/or disability (via the Original Reason for Entitlement Code). CMS is continuing to work on enhancements to the measure.