The Centers for Medicare and Medicaid Services has changed how the long-stay antipsychotic medication quality measure is calculated so that all residents with a history of schizophrenia are excluded from it, effective with the April 2026 refresh of the Five-Star ratings. The agency's own Design for Care Compare Nursing Home Five-Star Quality Rating System: Technical Users' Guide describes the change as bringing the calculation into line with established specifications, states that a subset of those residents had not been appropriately excluded from the measure data released in January 2026, and says the correction did not have a significant impact on scores for the measure or on quality measure ratings. The guide adds that scoring cut points for the measure have been reset to place providers into deciles using data from the fourth quarter of 2024 through the third quarter of 2025.

The exclusion itself is not new: residents recorded with schizophrenia, Tourette syndrome or Huntington's disease have long been kept out of the denominator, on the reasoning that antipsychotics are indicated for those conditions. What is new is the timing. On 19 March 2026 the HHS Office of Inspector General published an issue brief, OEI-02-23-00201, drawing on 40 nursing home surveys conducted between 2018 and 2021, which concluded that homes inappropriately assigned schizophrenia diagnoses to residents on antipsychotic drugs in order to improve their ratings on Nursing Home Care Compare. OIG recommended that CMS build on its work to reduce inappropriate schizophrenia diagnoses; in its published response CMS asked that the recommendation be closed as implemented and noted that the findings come from surveys predating its recent changes.

The more consequential change may be the one made three months earlier. From the January 2026 refresh CMS replaced the old measure with a respecified version that adds Medicare and Medicaid claims data and Medicare Advantage encounter data to Minimum Data Set records. In memorandum QSO-25-20-NH, revised 10 September 2025, the agency said the extra data would capture antipsychotic prescribing that falls within the nursing home stay but outside the seven-day MDS look-back window, and would also be used to validate the schizophrenia, Tourette and Huntington's exclusions reported on the MDS. CMS projected that the national percentage of long-stay residents recorded as receiving an antipsychotic would rise from 14.64 per cent under the old measure to 16.98 per cent under the new one. Claims validation therefore runs in both directions: it can strike out an exclusion the MDS asserts, and it can confirm a history the MDS records.

What the guide does not supply is a count. CMS has not published how many residents the April correction moves out of the denominator, how many MDS-reported exclusions the claims check has overturned, or how the two effects net out at facility level. For families, the practical point is narrower than the policy row: the measure describes what a home reports, not whether a particular prescription is right for a particular person. Antipsychotics in dementia carry known risks and are prescribed and reviewed by clinicians, not by a rating system.