What changed
CMS published version 18.0 of the MDS quality-measure manual for January 1, 2026 and later issued an errata clarification for the long-stay antipsychotic measure. CMS had previously announced that claims and Medicare Advantage encounter data would supplement MDS data and help validate exclusions.
That makes comparisons across the specification boundary hazardous. A different value does not, by itself, prove a corresponding change in prescribing or resident care.
Read the rate with four labels attached
Every internal report should identify the measure version, observation period, data source, and exclusions. Without those labels, a facility can spend time explaining a movement created partly by the calculation rather than the underlying workflow.
Use the population measure to locate a question, then return to resident-level review for clinical interpretation. The measure is not a substitute for indication, symptoms, risks, goals, monitoring, and prescriber judgment.
Keep operational and public-reporting views distinct
A consultant's current worklist may use fresher pharmacy data than a lagged public measure. Both can be useful, but they answer different questions. Label them so a facility leader does not assume they reconcile in real time.
When software vendors advertise benchmarking, ask them to name the specification, refresh cadence, exclusions, and source data. A chart without those details is not a comparable CMS measure.